SITALWeek #426

Welcome to Stuff I Thought About Last Week, a personal collection of topics on tech, innovation, science, the digital economic transition, the finance industry, and whatever else made me think last week.

Click HERE to SIGN UP for SITALWeek’s Sunday Email.

In today’s post: Google's Gemini is out and it has some advantages over ChatGPT4, including being able to act as your phone assistant; the absurdity of accelerating spending on leading-edge chips by an order of magnitude; some ramblings on virtual reality; the potential to create virtual patients for clinical trials; The Curse and Falling Up; the youngest Boomers turn 60 as childless couples grow their wealth; unofficially, China is still by far the largest US trade partner; and much more below. 

Stuff about Innovation and Technology
AI Election Influencers Have Left the Barn
The FCC has outlawed robocalls with AI-generated voices in an attempt to prevent election influence. The real AI-influencer election battle will, however, take place on social media and via private messaging apps – and thus largely play out beyond the reach of regulation. The next US election contestation may be pinned on claims that people voted “under the influence” for a candidate they otherwise would not have supported. 
 
Leading-Edge Chatbots: N=2
Google’s Gemini Advanced launched last week. It’s a souped-up version of my favorite chatbot (formerly known as Bard), and I recommend people check it out if you haven’t already. If you have a Pixel phone, you should also be able to download the new Gemini app from Google, and then you can set it to be the default assistant on your phone (I am finding that to still be a bit buggy). Ethan Mollick has a good overview here that concludes: 
Gemini shows that Google is in the AI race for real, and that other companies besides OpenAI can build GPT-4 class models. And we now know something about AI that we didn’t before. Advanced LLMs may show some basic similarities in prompts and responses that make it easy for people to switch to the most advanced AI from an older model at any time. Plus, GPT-4’s “spark” is not unique to OpenAI, but is something that might often happen with scale. We don’t yet know if models get “sparkier” and more AGI-like as they get larger, but I suspect we will find out.
As I've noted in the past, I believe chatbots will be the primary user interface going forward, replacing smartphone and other operating systems.
 
Altman’s Trillions
Sam Altman reportedly plans to build out semiconductor manufacturing capacity to fuel the AI revolution. According to the WSJ, Altman is looking to raise as much as $5-$7 trillion (that’s trillion with a "T"), to build factories to churn out GPUs and other AI-focused processors. For context, the entire semiconductor industry spends ~$150B/yr in capex, with a little more than half of that going to equipment. Altman’s plans, even if he deployed that capital over ten years, would be multiples more than the current entire semiconductor industry (not to mention, the $150B supports diverse end markets beyond AI including memory, analog, microcontrollers, etc.). Said another way, Altman wants to spend somewhere in the neighborhood of 20x TSMC's annual capex of ~$32B, which is a relevant comparison given TSMC builds almost every AI processor globally. Without any additional growth, the chip industry is already looking at lead times of 12-18 months for key pieces of fab equipment, like lithography machines from ASML. Altman’s plan would likely call for the purchase of 10x as many of these machines as the rest of the industry combined. We’ve previously discussed the friction encountered when interfacing with the slow analog gears of the economy, and manufacturing necessary hardware is only one such example. There are further gating factors such as engineering availability, an issue that has slowed TSMC’s global expansion to date. Last week, I speculated on whether a business model existed that could support the billions in incremental capex on AI; however, I don’t believe I currently have the imagination to underpin trillions of dollars. While it's likely impossible to expand chip capacity to that extent, I think it's more likely AI will create many ways to make itself require less processing power over time. If we see a scenario where the demand is there for 10x the number of chips, but those chips cannot physically be made, then AI would likely be reserved for the people or companies with the most money. Thus rendering popular tools like GPT4 or Gemini unavailable to the masses.
 
Bringing the Virtual to Reality
Disney’s $1.5B investment in Epic to bring Disney content to the Fortnite universe is underpinned by the game-maker’s Unreal engine. Disney’s cinematic directors were early users of Unreal. Even as far back as 2020, Jon Favreau was using the gaming engine to create The Mandalorian. As I wrote last week, we seem quite close to being able to create virtual worlds that are persistent and interactive. Integrating Unreal, AI, and VR/AR into the movie-making process at Disney could allow future viewers to experience their own narratives. This idea – ripped from the pages of Snow Crash and Ready Player One – was once called the metaverse, but that term slipped out of style after a failed hype cycle a couple of years ago. Regardless of what you call it, we will soon experience a fundamental shift in human storytelling as creators and artists move from linear, 2D thinking to full spatial computing and immersive entertainment. The potential addictiveness of this new type of entertainment is frightening. I am hopeful that current VR goggles morph into all-day wearable augmented reality glasses – I’d much rather bring AI/virtual elements into reality than transform reality into something entirely virtual. 
 
That sentiment – my hope/optimism for mixed reality triumphing over complete VR – brings me to the latest post from my favorite technological essayist, Jaron Lanier, who recently penned a reaction to the Apple VR goggle launch in The New Yorker. Lanier has a terrific insight on why VR gaming hasn’t taken off like many (including myself) thought it would: “There are many reasons why V.R. and gaming don’t quite work, and I suspect that one is that gamers like to be bigger than the game, not engulfed by it. You want to feel big, not small, when you play.” Lanier also expects the killer apps will yet land for mixed reality, and he believes they will either be practical, short-term, and task-based (“a scattering of wonderful niches”), or truly wild and weird, such as mapping your body to that of a lobster or playing with the flow of time (yes please!). Whatever it is, it’s unlikely that VR’s killer app will just replicate what we do on phones/laptops via goggles that isolate us from the rest of the world. Lanier offers his views on technology and our relationship with it:
It is impossible to judge technology without a sense of its purpose—and its only plausible purpose is to benefit people, or perhaps animals, or the overall ecosystem of the planet. In any case, if we pursue technologies that make it hard to delineate the beneficiaries—for instance, by blending brains into robotics not to cure a disease but just because it seems cool—then we make the very idea of technology absurd. The central question of the technological future is how to identify the people who are supposed to benefit from technology, especially if they seem to have melted into it. If people aren’t special, how can we act in a way that benefits people? We can’t. The principles of ethics, design, and even technology itself become nonsense. What can that specialness be? It must be something that is not technologically accessible, since technology expands unpredictably. It’s a little mystical. The definition of people must be one of apartness. We must now put people on pedestals, or they will drown...
Lanier also wonders – after remarking on the relevance of the (predominantly young male) “fantasy of infinities” that dominates so many visions of new technologies – whether VR might finally break humanity:
And so another urgent question is whether people can enjoy the storied reality of finitude after coming down from the high of fake infinity. Can being merely human suffice? Can the everyday miracle of the real world be appreciated enough? Or will the future of culture only be viral? Will all markets become Ponzi-like fantasies? Will people reject physics forever, the moment we have technology that’s good enough to allow us to pretend it’s gone?
To conclude, here’s one more philosophical note from Lanier: “Life within a construction is life without a frontier.”

Miscellaneous Stuff
Clinical Twins 
With the potential for significantly accelerated drug discovery as AI advances in capabilities, a major gating factor will be the analog process of conducting clinical trials. One company that recently raised venture capital is Unlearn, which “uses machine learning to create digital twins of clinical trial participants before being randomized in a controlled trial. The digital twin can provide researchers insight into the participants' health outcomes. The startup pitches its technology as a way to run smaller clinical trials more quickly, since researchers can find fewer participants for the control group.” (As participants’ digital twins will also serve as controls). I’ve been skeptical of the digital human twin concept in the past given how complex the body is, but it is possible such AI-simulated controls will be able to fast-track the process of identifying which drugs have the best shot at success in larger clinical trials. 
 
Falling Up
I’ve been wanting to write about The Curse (streaming on Paramount+) since I saw the first episode, but I patiently waited until the final episode aired a few weeks ago to try and collect my thoughts. However, I’m still struggling a bit to piece my views together. It’s certainly the unusual show we expect from the minds of Benny Safdie and Nathan Fielder. I’ve previously written about Fielder’s reality-bending creations (see Analog VR), noting his previous show, The Rehearsal, “is ultimately like an analog version of VR – an attempt to simulate a complex situation in realistic detail.” In some ways, I cannot really recommend anyone watch The Curse unless you want to put yourself through the same torture of trying to comprehend it all. Director Christopher Nolan recently called it an “incredible” show “unlike anything I’ve ever seen on television before”, declaring that the “tone is the star” of the show (not unlike Twin Peaks in the early 1990s). The purported concept of the show is a behind-the-scenes look at a couple trying to launch a HGTV home renovation show, but that’s a simple pretense for much more. I’ve read a lot of reviews and watched all of the creators and actors walk through every episode in various interviews, and there are as many interpretations as there are opinions on it. Director and co-creator Safdie responded to questions about the show’s ending by saying: “the way that I'd like to leave it is if you have a drawer, for example, and there was no handle on the drawer how would you open it? You know, would you tip it forward so that the drawer falls out? Would you use a knife to get into the crack and open it? Maybe you'll take some tape and stick it to the front and then pull it out that way. You know, you'd come up with a lot of different ways to open that drawer and what you'd find inside is the same information, but the way that you got that information was your own. If you put a handle on the drawer you just open it right up and you find your information so you lose a little bit of that process and that kind of searching which I think is important for this.” (This quote comes from an interview discussing the season finale, so if you have not watched the show and plan to, I recommend not watching that interview yet; likewise, I won’t discuss the fascinating character studies and perversions of behavior that lead up to the finale.) What attracts me to the show, despite its often unbearably uncomfortable plot/characters, is the same theme I am repeatedly drawn to: playing with reality. It seems that the human brain experienced a paradigm shift with the advent of the motion picture, as we could, for the first time, stare at someone acting out real situations without them knowing or staring back at us. This element of voyeurism is at the heart of reality TV, and it’s certainly a key element of The Curse. It’s one thing to watch reality TV, it’s an entirely different layer of the onion to spy on the lives of reality-TV creators. The Curse is filmed in such a way that you, the viewer, feel as if you’re entirely concealed so that no one can watch you stare and react. It’s truly like watching someone who has no idea they are being watched. As I concluded in Analog VR: “That is precisely what VR/AR will allow: realistic simulations of real-life situations with seemingly real people played by AI actors.” This is in part what The Curse felt like to me – a glimpse at our ability to eavesdrop on reality as curious voyeurs, and it’s directly tied to the above section on Disney’s investment in Epic and Lanier’s comments on VR. There’s a children’s poem I can’t get out of my head since I finished the show. It’s Falling Up by Shel Silverstein:
I tripped on my shoelace
And I fell up–
Up to the roof tops,
Up over town,
Up past the tree tops, 
Up over the mountains,
Up where the colors
Blend into the sounds.
But it got me so dizzy
When I looked around, 
I got sick to my stomach
And I threw down.

Stuff About Demographics, the Economy, and Investing
Demographic Surf Report
The youngest Baby Boomers turn 60 this year, which brings the generation to the cusp of their peak retirement numbers. This steady loss of workforce is no doubt contributing to the low unemployment rates, as retirements outnumber new labor market entrants. The rise in teen employment I mentioned last week is a helpful buffer, on the margin. Once retired, consumption tends to decline, which could weigh on the overall ability for the US economy to grow. Yet, increased demand for healthcare and household assistance will be a big driver of demand for automation, robotics, and AI in the coming years. While we might expect Boomers to steadily sell stocks (people over 55 own 80% of US stocks) and strain Social Security as it pays out more than it takes in from the generationally smaller working-age populations, we also need to consider the extreme amount of wealth the Boomers have accumulated in home equity and other assets. It’s possible they will remain significant drivers of consumption across the economy. The other big impact on overall economic growth comes from Millennial and Gen Z couples. BI reports that although couples without kids (known as DINKs – double income no kids) are growing faster than couples with kids, couples of all types are seeing their net worths rise at a faster pace. Although, historically, families have driven an outsized portion of consumption, today’s DINKs have net worths over 50% higher than couples with kids. There’s typically never any strong conclusions from demographics, and the above stats highlight how complex the interplay of intergenerational trends can be. However, there are likely some multi-year currents to ride based on these shifting age groups and family dynamics, so let me know if you spot a good wave out there.
 
US Imports: Mexico > China?
The US government’s import stats suggest that imports from Mexico surpassed those from China for the first time in 20 years. While imports from Mexico were only up modestly, imports from China dropped 20%. According to the NYT, the drop is largely due to a decline in consumer appetite for Chinese manufactured goods following excess pandemic demand, although the level of imports is below where it was pre-pandemic, and, given inflation, the real volume might be even further below those levels. However, these “official” numbers appear to ignore the de minimis imports of under $800 in value. As I mentioned in last week’s post on Shein and Temu, this is a large and growing piece of trade with China. The under-the-radar imports could be nearly $200B, and if you assume a meaningful portion of these would qualify as goods imported for purchase in the US, then I suspect China is still by far the largest US trade partner.

✌️-Brad

Disclaimers:

The content of this newsletter is my personal opinion as of the date published and is subject to change without notice and may not reflect the opinion of NZS Capital, LLC.  This newsletter is an informal gathering of topics I’ve recently read and thought about. I will sometimes state things in the newsletter that contradict my own views in order to provoke debate. Often I try to make jokes, and they aren’t very funny – sorry. 

I may include links to third-party websites as a convenience, and the inclusion of such links does not imply any endorsement, approval, investigation, verification or monitoring by NZS Capital, LLC. If you choose to visit the linked sites, you do so at your own risk, and you will be subject to such sites' terms of use and privacy policies, over which NZS Capital, LLC has no control. In no event will NZS Capital, LLC be responsible for any information or content within the linked sites or your use of the linked sites.

Nothing in this newsletter should be construed as investment advice. The information contained herein is only as current as of the date indicated and may be superseded by subsequent market events or for other reasons. There is no guarantee that the information supplied is accurate, complete, or timely. Past performance is not a guarantee of future results. 

Investing involves risk, including the possible loss of principal and fluctuation of value. Nothing contained in this newsletter is an offer to sell or solicit any investment services or securities. Initial Public Offerings (IPOs) are highly speculative investments and may be subject to lower liquidity and greater volatility. Special risks associated with IPOs include limited operating history, unseasoned trading, high turnover and non-repeatable performance.

SITALWeek #425

Welcome to Stuff I Thought About Last Week, a personal collection of topics on tech, innovation, science, the digital economic transition, the finance industry, and whatever else made me think last week.

Click HERE to SIGN UP for SITALWeek’s Sunday Email.

In today’s post: the Zoltar arcade machine gives us some insight into the billions upon billions being poured into AI development; the Chinese fast-fashion loophole that is fueling the digital advertising industry; the afternoon snack war; layoffs are low as employment remains high; the widening political gender gap; and much more below. 

Stuff about Innovation and Technology

“Your Wish Is Granted”
Lately, you can’t interact with a news algorithm without hearing something about another $10B+ AI spending plan. Alphabet spooked investors last week by their reported spending of $11B in capex for the fourth quarter – driven “overwhelmingly” by “technical infrastructure” – and their expectation that capex will be “notably larger” in 2024. Google started the year with $6.3B in Q1 and steadily ramped to the monster sum of $11B, giving a total of $32B in capex for 2023. Just using the Q4 run rate would imply ~$44B for 2024, or around 15% of the revenue that analysts expect the company to earn. Meanwhile, Microsoft reported a similar $10.7B in capex including finance leases for Q4, up over 20% from the prior year. We’ve also learned, in an interview with Mark Zuckerberg I referenced two weeks ago, that Meta plans to buy billions of dollars of GPUs just this year. And, OpenAI’s CEO Sam Altman is rumored to be raising billions to set up joint ventures with chip makers like TSMC to help fund chip fab. I suspect this maneuver is largely out of fear that OpenAI won't be able to secure capacity at TSMC fabs for GPUs (or, potentially, their own custom AI chips). Elon Musk, for his part, is rumored to be looking to raise $6B for X.AI, which seems almost adorably quaint given how little progress that would fund relative to the spending intentions and deep wallets of the competition. Trillion-dollar investment giant Blackstone has $15B of new data centers in development (in somewhat of a circular reference) as they buy into the capex hype. While I’ve speculated in the past that a breakthrough in efficiency, developed by AI itself, could decrease the need for leading-edge chips, that appears to not be on the immediate horizon. 
 
All of these speculative billions upon billions of dollars in expenditures raise the obvious question: what business model is going to pay for it all? Hundreds of billions were spent to build the Internet; but, large-scale business models to justify it didn’t show up until around a decade after the dotcom crash. Currently, the leading-edge chip industry (and many trailing-edge chips) are largely paid for by two industries: advertising and cloud software. Behind the ads lie a variety of entertainment choices (e.g., video, social networks, games, etc.). I am of course generalizing here, but the vast majority of use cases for leading-edge chips are phone processors, servers, GPUs, networking, wireless connectivity, etc. Digital ads (including those funding search and chatbots) and cloud software are each multi-hundred-billion-dollar industries that (in yet another circular reference) are creating even more demand for leading-edge chips. When I look back at my post titled End of Advertising-Funded R&D? from November of 2022, it appears I was overly pessimistic about the prospects for the next wave of new markets, despite my enthusiasm at the time for AI transformer models and AR.
 
Let’s ignore for the moment the euphoric spending cycles early in new tech platforms that lead to bubble formation and speculate on what might sustainably pay for hundreds of billions of dollars of new chip capacity. First, I think we can safely say that the advertising (entertainment) and enterprise software industries will be heavy users of AI, and this usage will drive higher revenues for all of these segments. We can now see a clear, near-term path to holodeck-like world creation with simple text prompts. Such a feat will require a massive amount of chip investment and new leading-edge hardware; and, if it’s as fun as my VR boxing app, people and advertisers will be happy to pay for it. When you look at the progress that Google is making with projects like Lumiere, it’s not a leap to imagine how much processing power we will need. Further, AI is going to steadily replace and/or augment all white collar jobs to such an extent that companies will be more than happy to pay up for their new AI employees.  
 
Robotics is another example of a market that could grow into the trillions of dollars in the coming decades with a variety of special purpose and general humanoid form factors powered by AI. Our new AI-enhanced assistants will be massive consumers of chips for both the hardware and cloud processing necessary for navigating our complex world. 
 
Now, let’s go out a little further on the limb and speculate what might really be behind this euphoric FOMO GPU land grab. What’s the real pot of gold that these billionaire tech CEOs might be after? My candidate is: predicting the future. I discussed this in more detail in my longer post titled Simulacrum a few months ago. Effectively, AI is going to give us the ability to simulate countless realities in parallel, and, in doing so, will create a new type of modeling and virtual forecasting. Remember when sports books were using video games of professional teams programmed with player stats to gauge who might win real-world games? Now expand that concept until you get to a “game” that predicts the entire world on a day-to-day (or longer-term) basis (or, perhaps you might not remember this, because a quick conversation with Google Bard leads me to believe that video games were never used to calculate gambling odds!). Ultra-powerful AI holds the promise of a real Laplace’s Demon with perfect knowledge of the past, present, and future (for more on this thought experiment, check out the TV show Devs on FX/Hulu). When Nvidia announced its Earth-2 simulator in 2021, the blog post was titled “...to See Our Future”. If you had a slight edge in knowing, for example, what type of unexpected and erratic behavior could create a new type of political candidate, how much would that be worth? The benefits (financial and otherwise) of being able to predict anything, from what therapeutic drugs might work to quarterly earning stats, are obvious. Of course, all of this goes against the fundamental tenets of the biggest foundational belief we have: the world is a complex adaptive system with emerging behaviors and wildly unpredictable outcomes based on chaotic initial conditions that are constantly evolving. So, the question is, with enough chips, could you overcome the limitations of living in a complex adaptive system – with googols of variables – to virtually time travel into the future? Who hasn’t wanted their own Zoltar to grant your wishes? If you thought there was even a small chance of creating such a machine, how much would you spend to be the first one to do it?
 
Algorithmic Fashion
The US textile industry is trying to fight back against the small-package loophole exploited by the Chinese apps Shein and Temu to take over the US fashion industry. The rise of these prolific peddlers of disposable clothing (and other trash, trinkets, and whatever the heck those weird clickbait objects are in all of their ads!) have disrupted industries and brought a windfall of billions of advertising dollars to companies like Meta. Temu is owned by the Chinese public company PDD, while Shein is looking to IPO in the US. Shein reportedly uses algorithms to automatically identify and create hot clothing items based on TikTok trends and the like with little to no human involvement. The online discount apparel trend is impacting the fast-fashion giants and the discount apparel retailers in the US, many of which lack a viable online strategy and certainly aren’t using AI algorithms to instantly produce knockoff goods. Even brands like Gap and others are blaming Shein for cutting into profits. Curiously, the Chinese fashion apps are more popular with Gen X and Boomers. The trade loophole being exploited by Shein and Temu, according to the NYT, is called de minimis and it allows anything under $800 to slip by without tariffs. More typically, a retailer would place a bulk order from overseas (large enough to require paying tariffs), and then break it up to ship to stores or ecommerce customers. Congress identified the potential issues with the law in a 2023 report titled “Fast Fashion and the Uyghur Genocide” (page 2 of this PDF covers the key points). I don’t have a view as to whether Shein and Temu are sustainable business models. I will say that it’s hard to get too excited about companies that need to spend billions of dollars on social networking ads to draw customers and still haven’t created a lawful, legitimate business model (Chinese companies, no doubt led by Shein and Temu, accounted for 10% of Meta’s revenues and 5% of their 16% y/y growth, according to their last quarterly conference call). But, the concept of algorithmic driven fashion and, more broadly, AI-driven retail purchasing is more interesting. TikTok last week announced that any video will be shoppable using AI to identify anything and everything for sale. The speed at which fashion moves from smartphone screen to closet makes your head spin.
 
Starbucks Snacking: GLP-1 Casualty? 
Starbucks is claiming to be the victim of an afternoon snack war as they see fewer visitors at that time of day. Although it appeared to not surface on the company’s earnings call last week, I can’t help but wonder if their real enemy in the afternoon snack war is the GLP-1 class of weight loss drugs? In related news, STAT reports that researchers are looking into whether GLP-1s can also treat depression and anxiety: “Early data and anecdotes suggest that this class of GLP-1 drugs could help patients feel less depressed and anxious. The treatment may also fight the decline in cognitive and executive function that many people with mental health disorders experience, like worsening memory and losing the ability to focus and plan.” Of course, the reason GLP-1's were created was because of the afternoon snack wars…

Miscellaneous Stuff
Fermenting Plastic
Researchers have engineered a bacterial strain to upcycle plastic into a biodegradable “spider silk”, which could have a host of commercial applications. The report is only a preliminary proof-of-principle, but it offers an interesting option for untangling ourselves from our ever-accumulating polyethylene trash. 

Stuff About Demographics, the Economy, and Investing
Surprisingly Strong Employment Data
Despite fears of AI-driven layoffs and a seemingly never-ending string of press releases about large corporate downsizings, the macro data imply that we are below trendline, with around 1% discharges in 2023 vs. 1.25-1.5% for the previous decade, according to a BI story. The tech sector is closer to its long-term average rather than above it. The big issue may not be layoffs but a lack of new job creation, which has been anemic for the IT sector (see AI Layoffs). One generation that’s working a lot harder than before is Gen Z. The WaPo reports that teens with jobs climbed from a low of 34% in 2014 to 37% last year. I recall that I and my fellow Gen Xers were all labeled slackers, but that’s just because we were a very small birth cohort – there weren’t enough of us to fill all the afterschool, weekend, and summer jobs to serve the Boomers! It turns out, depending on the year, Gen X had 40-50% working teenagers (similar to Boomers) making the real slackers Millennials and Gen Z, despite the recent uptick. The increased youth workforce is largely functioning as a pressure-relief valve for understaffed businesses like restaurants. 
 
Political Gender Gap
I tend to read reports that classify generalized trends across populations with a great deal of skepticism, but it’s hard not to take notice of recent data purporting a liberal/conservative gap widening between men and women, as, anecdotally, the trend feels somewhat real. According to data from Gallup (as reported by BI), over the last decade, the political gender gap widened from around 5% to over 15%, with women going from 30% to 42% liberal and men staying around 20% liberal. Kurt Vonnegut once said: “Thanks to TV and for the convenience of TV, you can only be one of two kinds of human beings, either a liberal or a conservative”. Now, presumably thanks to social media’s digital tribalism, those polarizing terms are being subject to genderfication. 

How to Have Meetings That Don't Suck
In case you missed it, check out our new whitepaper on how we approach meetings at NZS Capital in a way that we actually get value out of them.

✌️-Brad

Disclaimers:

The content of this newsletter is my personal opinion as of the date published and is subject to change without notice and may not reflect the opinion of NZS Capital, LLC.  This newsletter is an informal gathering of topics I’ve recently read and thought about. I will sometimes state things in the newsletter that contradict my own views in order to provoke debate. Often I try to make jokes, and they aren’t very funny – sorry. 

I may include links to third-party websites as a convenience, and the inclusion of such links does not imply any endorsement, approval, investigation, verification or monitoring by NZS Capital, LLC. If you choose to visit the linked sites, you do so at your own risk, and you will be subject to such sites' terms of use and privacy policies, over which NZS Capital, LLC has no control. In no event will NZS Capital, LLC be responsible for any information or content within the linked sites or your use of the linked sites.

Nothing in this newsletter should be construed as investment advice. The information contained herein is only as current as of the date indicated and may be superseded by subsequent market events or for other reasons. There is no guarantee that the information supplied is accurate, complete, or timely. Past performance is not a guarantee of future results. 

Investing involves risk, including the possible loss of principal and fluctuation of value. Nothing contained in this newsletter is an offer to sell or solicit any investment services or securities. Initial Public Offerings (IPOs) are highly speculative investments and may be subject to lower liquidity and greater volatility. Special risks associated with IPOs include limited operating history, unseasoned trading, high turnover and non-repeatable performance.

SITALWeek #423

Welcome to Stuff I Thought About Last Week, a personal collection of topics on tech, innovation, science, the digital economic transition, the finance industry, and whatever else made me think last week.

Click HERE to SIGN UP for SITALWeek’s Sunday Email.

In today’s post: are EVs a bad deal? If you haven't tried boxing in VR, I highly recommend it  and, you don't need to spend $3,500 to do it; a Disney Imagineer creates a floor for the holodeck and VR; I journey to the weird world of talismans and 1980's kids programming and contrast it with today's algorithmic cookie cutter content; the JWST is causing us to rethink cosmological theories; immigration is roaring back in the US, and that's a good indicator for the economy. 

Stuff about Innovation and Technology
EV Chill
Car rental firm Hertz made headlines recently when it declared it was shedding 20,000 Teslas after disappointing results following their 2021 purchase contract (which was for a planned 100,000 vehicles). While some of Hertz’s issues won’t apply to the average consumer, it is worth looking at the problems and costs they’ve experienced. Many of Hertz’s EVs were purchased for use by Uber drivers; however, the high utilization/mileage of rideshare vehicles caused elevated wear and tear as well as an increase in minor collisions, which are more expensive to repair for a Tesla than the average rental car. EVs also have higher purchase prices and higher depreciation (losing 49% of their value after five years vs. 39% for gasoline cars). The higher costs to purchase and repair also inflate insurance rates. The incremental 10% depreciation on a ~$40,000 car would likely require Hertz to expense an extra $800/year. I suspect the increased depreciation is due to a belief that battery packs lose value faster than gasoline engines, but I think that fear is largely overblown, making used EVs a relative bargain. For a typical consumer, these higher costs are in part offset by lower maintenance; but, for a high utilization owner like Hertz, the costs appear to exceed the benefits. When Hertz moved some Teslas from the rideshare to the consumer rental fleet, many consumers didn’t want the extra hassle of dealing with charging on a vacation or business trip. As a result, Hertz is now de-emphasizing EVs. Uber, for their part, is still focused on their carbon footprint and electrifying the cars their drivers use for ridesharing with $2,000 purchase incentives and a partnership with Tesla to map out better Supercharger locations. 
 
The recent cold weather also made headlines and headaches for EV drivers, with many finding that their vehicles wouldn't even charge in subzero weather. Part of that is simply an element of preparedness; just like it makes sense to have gas or emergency provisions, EV owners should charge up ahead of storms (and/or make sure they have a battery warmer). However, all of these problems underpin the leveling off of growth in EVs in the US, which has some unique driving demands (e.g., more rural mileage, scarcity of chargers, etc.) relative to other areas of the world. I have long held that the far superior path to greening up our mileage is to mandate that every car have a small hybrid battery pack that would go 20-30 miles. Such a requirement would likely result in well over 50% of miles driven being powered by electricity vs. gasoline. PHEVs (plug-in hybrid electric vehicles) also extend the longevity of the car given the relatively low miles that get driven on the gas engines. We’ve had a PHEV vehicle for almost six years, and over 60% of its ~60,000 miles have been battery powered, despite having a range of only 30 miles. If we shifted focus from long-range to short-range capabilities, we should be able to power more than half the accrued miles with electricity with only modest battery capacity additions (~1/10th that of long-range EVs). Once we’ve tackled the majority of miles driven with short-range batteries, hopefully technological advances and cost declines would then allow for a complete shift to electric, assuming that's still the leading tech in the future. An interesting concept is the 2025 Ramcharger truck, which features a relatively small 92KWh battery pack (~100-mile range) with a 130KW gas-powered generator. Thus, rather than needing separate gas and electric drivetrains, the truck’s single electric motor is powered on the fly by the generator once its range is expended.
 
VR Boxing Match
One of my favorite recent experiences is the Supernatural boxing/fitness app on the Meta Quest 3. I’ve been a bit too skeptical about full VR headsets (and I still stand by my view that AR glasses will win out over passthrough VR goggles in time); so, I forced myself to use the Quest 3 daily for several weeks. I ended up liking the workout apps the most. Now that I’ve said something nice about the $500 Meta goggles, I think I can express a bit of skepticism, bordering on cynicism, over Apple’s $3500 Vision Pro. The Quest already has a healthy app developer ecosystem, whereas many of the top-ten apps on iOS are so far choosing to shun Apple’s VR hardware (perhaps in retaliation for Apple charging app developers exorbitant tariffs for decades!). This feels less like an iPhone vs Blackberry moment (when nearly everyone failed to see the astronomical value in a platform shift to multi-touch screens), and more like a McLaren vs a Toyota Camry, i.e., both are going to get you to the same place in roughly the same amount of time, but one offers far more value. In other words, a fancier and more expensive set of goggles may have little impact on your VR experience. While it’s early days for VR, my experiment with the Quest has greatly increased my optimism around the potential for VR and AR (and I was already quite optimistic about mixed reality). I also agree with Zuckerberg that the big leap in AR/VR is going to be powered by AI-created virtual worlds. The next thing I want to purchase to go along with my Quest is this very cool HoloTile floor created by a Disney Imagineer.

Miscellaneous Stuff
Long Lost Original Content
I read with some intrigue this article from a BI reporter that used ChatGPT to track down an old TV show from their childhood with only vague imagery of a “surreal scene where a massive toddler watched the miniature cast run around.” Who among us hasn’t tried to figure out “what was that movie where the kid put peanut butter on his head and his hair wouldn’t stop growing?” I had my own mystery show from the 1980s. It revolved around a bizarre post-apocalyptic world with spacey music, dark imagery, and the Dewey Decimal System. While the weirdness of the show and the haunting theme song are what stuck with me, I also had a vague memory that it was intended to teach elementary school children how to use the library. It’s been years since my last unsuccessful attempt to identify the show, so I too set out on an LLM-assisted quest. I worked with Bard (which I still much prefer to ChatGPT for these types of queries) and quickly discovered the 13-episode Tomes and Talismans series, which was created in 1986 by the Mississippi Public Broadcasting network. In 2020, the MPB posted the episodes online, and, in December of last year, they made them permanently available to viewers. While I remember very little of the plots/characters, the theme song was a positive match and confirmed that Bard had identified the correct show (I have yet to track down the composer). Here is a description of the show from MPB: “Atmospheric pollution and overpopulation were serious concerns on Earth by the end of the 21st century. Human movement was severely limited. The Wiper colonization of 2103 could not be stopped. This primitive species from the Dark Star solar system found Earth perfect for their favorite pastime: the disruption of all communication and data technology. The spread of Wipers and pollution increased the need for the scientific search for solutions. In 2117, the world-wide government voted for the gradual evacuation of all Earthlings to the White Crystal solar system. This report was completed in 2123 during our final days on Earth. We leave it for those who may return here in a storage vault for human learning...the last Earth library.” Frankly, that synopsis does not even begin to describe the strangeness of the series. MPB has created a page with lots of T&T content, and the episodes are also all on YouTube. If there is one takeaway from all of this, it’s that 1980s children’s programming was so wild and original that it haunts many of us decades later. It’s hard to imagine the algorithms that dictate whether pink pigs, baby sharks, or floating toilet heads are offered up as entertainment fodder could fish out, from today’s infinite sea of ephemeral content, anything as memorable as Tomes and Talismans, disincentivizing artists from creating anything with more than five minutes of staying power. It’s with some irony that I note the series was resurrected in late 2020 in no small part due to a post on YouTube from a vlogger that reminded many Gen X’ers of the series’ existence.

Cosmic Megastructures
The scientific method was a phenomenal advance, allowing us to come up with an idea, test it, and either reaffirm or change our mind based on the evidence. Observational results from space are giving us an opportunity to come up with new theories to test as old ones are disproven. The JWST has surfaced data – such as spotting more mature galaxies far earlier in the history of the Universe than expected – that upend several cosmological theories about the origin and evolution of the Universe. Further, data from the Sloan Digital Sky Survey recently revealed an unexpectedly large ring structure spanning 1.3B light years. Current theories would not have predicted the existence of such a structure, which encompasses over 1% of the known Universe. The so-called Big Ring is in the same neighborhood as the even larger Giant Arc (3.3B light years across) discovered in 2021. PhD student Alexia Lopez, who led the discoveries/analysis, commented: “These oddities keep getting swept under the rug, but the more we find, we’re going to have to come face-to-face with the fact that maybe our standard model needs rethinking. As a minimum it’s incomplete. As a maximum we need a completely new theorem of cosmology.”

Stuff About Demographics, the Economy, and Investing
Immigration Offset
According to Chinese official stats, there were 850,000 fewer births than deaths last year, suggesting that the population of the world’s second largest country is now in structural decline. The NYT covers failed attempts by other countries to boost birth rates by offering financial incentives and benefits like extended parental leave. In general, it appears such efforts, at best, pull in some births that people had already planned, resulting in a boom/bust cycle of births. As the world’s population ages and the elderly come to outnumber the youth, we are likely to experience demographically driven economic changes in the near future. I contemplated some of these in Jobless Economic Growth? a few weeks ago. Another topic worth thinking about is the wealth runoff of aging savers. Many boomer-aged folks are shifting to living off their retirement savings, possibly causing a steady selling of assets for income. And, we also know discretionary spending peaks when people are in their mid-50s and then starts to decline (this peak also tends to be for families, whereas people are now having fewer kids to spend money on). This combination of factors could create a demographic and spending headwind that countries should be vying to offset. As I’ve argued for several years, immigration remains GDP-growth’s best friend for developed nations, and policies that promote immigration should be a primary focus for any country looking to survive. Research from JBREC highlights the return of US immigration to pre-COVID levels, which is setting up the country for one of the better economic outlooks on a relative basis compared to other developed nations.

✌️-Brad

Disclaimers:

The content of this newsletter is my personal opinion as of the date published and is subject to change without notice and may not reflect the opinion of NZS Capital, LLC.  This newsletter is an informal gathering of topics I’ve recently read and thought about. I will sometimes state things in the newsletter that contradict my own views in order to provoke debate. Often I try to make jokes, and they aren’t very funny – sorry. 

I may include links to third-party websites as a convenience, and the inclusion of such links does not imply any endorsement, approval, investigation, verification or monitoring by NZS Capital, LLC. If you choose to visit the linked sites, you do so at your own risk, and you will be subject to such sites' terms of use and privacy policies, over which NZS Capital, LLC has no control. In no event will NZS Capital, LLC be responsible for any information or content within the linked sites or your use of the linked sites.

Nothing in this newsletter should be construed as investment advice. The information contained herein is only as current as of the date indicated and may be superseded by subsequent market events or for other reasons. There is no guarantee that the information supplied is accurate, complete, or timely. Past performance is not a guarantee of future results. 

Investing involves risk, including the possible loss of principal and fluctuation of value. Nothing contained in this newsletter is an offer to sell or solicit any investment services or securities. Initial Public Offerings (IPOs) are highly speculative investments and may be subject to lower liquidity and greater volatility. Special risks associated with IPOs include limited operating history, unseasoned trading, high turnover and non-repeatable performance.

SITALWeek #422

Welcome to Stuff I Thought About Last Week, a personal collection of topics on tech, innovation, science, the digital economic transition, the finance industry, and whatever else made me think last week.

Click HERE to SIGN UP for SITALWeek’s Sunday Email.

In today’s post: a day of reckoning dawns for the companies that mortgaged their futures to buy back shares over the last decade; 2024 is the year drones finally takeoff for ecommerce; the NFL has a broadcast problem; AI will multiply financial gains for artists; protecting humans from robots; IT jobs are no longer growing, and may shrink with AI; cool new flow batteries; Brett discusses Trane on Business Breakdowns; and a short lesson on applying complex adaptive systems in reverse.

Stuff about Innovation and Technology
SkyMart 
Drone delivery company Zipline is ramping its Walmart trials 1,000x to serve 1.8 million homes in the Dallas-Fort Worth metro area. Zipline makes a drone delivery every ~70 seconds globally and is approaching one million cumulative deliveries. This expansion of service is a significant development, which could give companies like Walmart a leg up over Amazon, whose drone efforts continue to lag far behind others. I've previously contemplated Walmart acquiring the assets of UPS or FedEx, which are struggling to grow volumes, but perhaps they can skip road-based delivery altogether. I wrote in more detail about Zipline in #389. 

Innovation Beats Buybacks 
As I read recently about Walgreens Boots Alliance's efforts to conserve cash by cutting its dividend and looking to sell off UK pharmacy Boots, I couldn’t help but return to a 2018 oped and a 2019 SITALWeek highlighting the risk of prioritizing stock buybacks and short-term shareholder returns at the expense of forward thinking innovation and customer problem solving. Walgreens announced a $10B share repurchase in 2018 around the same time Amazon announced it was acquiring PillPack to enter the pharmacy space. Clearly, I was wrong about many things back in 2018, including how little success Amazon has had in healthcare, but it seems obvious in hindsight that Walgreens neglected an opportunity to prioritize customer experience and invest in digital innovation. Since Walgreens announced the big repo five years ago, the stock is down nearly 50% while the S&P 500 is up over 90%. When disruption is on the horizon, as it is for nearly every company in every industry, businesses that evolve to meet their customers’ changing needs nearly always do far better than those that instead choose to return capital and sit on their hands. This is a good example of why we emphasize adaptability and non-zero-sum-maximizing companies at NZS Capital.

Lonely Linear NFL 
The NFL accounted for 93 of the top-100 broadcast network shows in 2023, up from 82 in 2022 and 61 five years ago. While the NFL continues to hold its popularity, these stats are far more indicative of the decline of all other broadcast viewing. The only two non-football related broadcasts in 2023 were the Academy Awards and the POTUS State of the Union (the other five slots were filled by pre/post football programming or college football). With significant challenges facing streaming platforms that rely on legacy cable and broadcast assets, it’s likely that it will be the NFL, more than any other force, that directs the future of these struggling companies (just as SITALWeek was going to press, the NY Post broke the news that the NFL is looking to take a stake in Disney’s ESPN). The rapid decline in linear-TV subscribers threatens networks' ability to make payments to the major sports leagues on a national basis like we’ve seen with regional sports rights (see Sporting Headwinds). The biggest threat to professional content still remains AI, with Hollywood becoming Vaudeville and holodeck-like technology just around the corner; however, a deteriorating backdrop for sports rights payments may be a larger, near-term threat.

AI’s Artistic Land of Opportunity 
Musicians have, on occasion, been on the cutting edge of financial innovation. Bowie Bonds pioneered the concept of performers burdening future royalty streams with debt in order to support their current lifestyle. In another example, music labels (and, to a lesser extent, the live music performance industry) act as specialized banks for lending to artists. ABBA’s virtual show added $225M to the London economy, and virtual KISS appears poised to be a likewise lucrative leveraging of AI. Taylor Swift is more of a corporation than a musician, a statement that in no way diminishes her creative abilities – many artists are also great business operators. The creators of South Park (and hit musical The Book of Mormon) recently upsized a $600M loan to an $800M loan that would include a cash payment secured against the future earnings of their intellectual property. These deals have come a long way: $800M is nearly 15x the size of the original Bowie Bond! With the advent of AI-developed music and virtual performances, there will be a proliferation of opportunities for recording artists to increase their income and for music rights holders to expand their value, assuming artists agree to the use of their voice/content. For example, I’d love to be able to hear any artist cover a song by any other artist in their own unique style. Who wouldn’t want to hear Billie Holiday cover Billie Eilish? Such an engine could be a lucrative amplifier of the value of music rights. While criticism and cynicism around AI’s impact on art of all kinds is rampant, AI will no doubt create far more opportunities than existential threats for artists. 

Embodied Ethics  
Google’s DeepMind is introducing an array of robotic platforms aimed at accelerating the development of general purpose robots leveraging AI. Over the course of seven months, Google ran 77,000 trials across 52 robots performing 6,650 tasks in its offices. Top of mind for Google is safety as robots increasingly interact with humans:  
It features safety guardrails, one of which is providing its LLM-based decision-maker with a Robot Constitution - a set of safety-focused prompts to abide by when selecting tasks for the robots. These rules are in part inspired by Isaac Asimov’s Three Laws of Robotics – first and foremost that a robot “may not injure a human being”. Further safety rules require that no robot attempts tasks involving humans, animals, sharp objects or electrical appliances. 
But even if large models are prompted correctly with self-critiquing, this alone cannot guarantee safety. So the AutoRT system comprises layers of practical safety measures from classical robotics. For example, the collaborative robots are programmed to stop automatically if the force on its joints exceed a given threshold, and all active robots were kept in line-of-sight of a human supervisor with a physical deactivation switch. 
There is a series of illustrations in the blog post announcing the new platforms. 

AI Layoffs 
Language learning giant Duolingo announced it was cutting 10% of its contract workers due to leveraging AI. This move is part of a broader trend of doing more with fewer people thanks to the new technology (see Artificial Efficiency for more). The entire US IT industry, which was one of the earliest adopters of AI copilots, saw virtually no growth in jobs in 2023 after adding 267,000 jobs in 2022, and engineers are finding the job market much more competitive.  

Finding Nano Flow 
A DARPA project has made meaningful progress on flow batteries, which can be refueled – by swapping out spent electrolyte fuel with fresh solution – in the same amount of time it takes to refill a gas tank (and the fluids can be reused roughly the same number of times that a lithium-ion battery can be recharged). Further, they cannot catch fire (the fluid is a fire suppressant), so extra protective infrastructure around the batteries is not needed (saving weight in an EV). Until recently, however, flow batteries lacked sufficient energy density to compete with lithium-ion batteries (because you can only dissolve so much electrolyte in solution, requiring prohibitively large electrolytic reservoirs to achieve long-range capabilities). Researchers at DARPA have overcome this limitation by using nanofluids, a technology that relies on particle suspension rather than dissolution, dramatically increasing the energy density of the electrolyte solution to rival that of traditional lithium-ion batteries. Given they can also be recharged from any electricity source, nanoflow batteries could also be a safer way to build a distributed grid with in-home energy storage. DARPA has an obvious military interest in such battery technology for use in combat vehicles, and I imagine this would also make sense for fleets and long-haul trucking. Like other types of novel battery technology, nanoelectrofuel batteries are a few years away from commercialization and suffer from the chicken/egg scaling problem common to new technologies that require broad deployment of supporting infrastructure, in this case charging/refilling stations. 

Miscellaneous Stuff
Doubling Drug Discovery 
DeepMind’s drug discovery startup Isomorphic Labs aims to halve the time it takes to discover new drugs, whittling the process down from five years to two. Focused on partnerships with Lilly and Novartis, the AI firm is eyeing up to $3B in performance-based incentives if they achieve their objectives. 

Stuff About Demographics, the Economy, and Investing
So Cool It’s Hot 
Our very own Brett Larson joined the Business Breakdowns podcast to discuss HVAC giant Trane Technologies. At NZS, we like to alternate between jokes about HVAC being “hot” but also “cool”. Additional interviews with NZS investment team members can be found here. 

Bidirectional Luck 
We often write about how complex adaptive systems teach us to stop falling into the trap of attempting to predict the future. But, we tend to forget that the arrow of prediction goes backward as well, namely in the form of attribution. We often believe we can attribute causation to coincident circumstances. In our own decision making, especially when it comes to investing, it’s all too easy to over-attribute negative outcomes to bad luck and positive outcomes to our own skill. In reality, luck plays a profoundly larger role than we often realize, both for attribution and prediction. We recommend doing pre-mortem analysis to make better forward-looking decisions, but we should also spend just as much time on counterfactuals to examine how things could have gone differently in the past. If we devoted as much time to showing how lucky we have been, it might help to create better circumstances for good luck to find us in the future.

✌️-Brad

Disclaimers:

The content of this newsletter is my personal opinion as of the date published and is subject to change without notice and may not reflect the opinion of NZS Capital, LLC.  This newsletter is an informal gathering of topics I’ve recently read and thought about. I will sometimes state things in the newsletter that contradict my own views in order to provoke debate. Often I try to make jokes, and they aren’t very funny – sorry. 

I may include links to third-party websites as a convenience, and the inclusion of such links does not imply any endorsement, approval, investigation, verification or monitoring by NZS Capital, LLC. If you choose to visit the linked sites, you do so at your own risk, and you will be subject to such sites' terms of use and privacy policies, over which NZS Capital, LLC has no control. In no event will NZS Capital, LLC be responsible for any information or content within the linked sites or your use of the linked sites.

Nothing in this newsletter should be construed as investment advice. The information contained herein is only as current as of the date indicated and may be superseded by subsequent market events or for other reasons. There is no guarantee that the information supplied is accurate, complete, or timely. Past performance is not a guarantee of future results. 

Investing involves risk, including the possible loss of principal and fluctuation of value. Nothing contained in this newsletter is an offer to sell or solicit any investment services or securities. Initial Public Offerings (IPOs) are highly speculative investments and may be subject to lower liquidity and greater volatility. Special risks associated with IPOs include limited operating history, unseasoned trading, high turnover and non-repeatable performance.

SITALWeek #421

Welcome to Stuff I Thought About Last Week, a personal collection of topics on tech, innovation, science, the digital economic transition, the finance industry, and whatever else made me think last week.

Click HERE to SIGN UP for SITALWeek’s Sunday Email.

In today’s post: I recap some of my favorite as well as some of the most popular posts from 2023. Before I dive into that you'll find a brief reflection on how to break free from the most common cognitive biases in order to improve decision making.

Over the holiday break, I came across a paper that attempted to reduce an array of cognitive biases to a small number of grouped behaviors. I tend to agree with the spirit of the paper, but my impulse was to reduce their reduction even further. We've focused on rooting out cognitive bias for decades (see chapter 6 of Complexity Investing), and it's a major focus of how we try to interact with each other and make investment decisions at NZS. At the most basic level, almost all of the hundreds of cognitive biases boil down to two things: 1) "knowing" that things in the future resemble things you "know" from your personal past, and 2) believing everything you come across fits with that "knowledge". Obviously, the problem is the illusion of "knowing" since, in reality, we are largely unprepared for the future, and what we "know" from the past is rarely as useful as we think it is.

There's clearly more nuance to cognitive bias, but I think the best way to improve decision making is to approach questions with a complete beginner's mind and strongly resist the urge to say "I've seen this before", no matter how much it might resemble something familiar. Complex adaptive systems have so many different variables and inputs that superficial resemblance to something you've seen before generally isn't useful. It's a challenging paradox for our brain. For every single investment I look at, I immediately start making countless comparisons to prior companies, industries, investment patterns, TV show plots, something that happened when I was 12...you name it, and I start comparing. Then I have to repeat the mantra "who cares!? it’s probably not relevant" and instead try to look at everything fresh. Having a beginner's mind strikes me as the hardest thing to do in investing and life. The roughest paradox of decision making is that the single best source of information we have at any given moment is what we already know, and yet it's a pretty lousy well to draw from.

The only practical advice I've seen on how to lesson our reliance on cognitive pattern-recognition ruts is from Lisa Feldman Barrett: "Your brain predicts (in large part) by reassembling your past experiences that are similar to the present moment. That means every new experience you cultivate for yourself – every new thing you read, every new person you talk to, every new thing you learn – is an opportunity to change what your brain will predict in the future, and which actions you may take. In other words, your brain (meaning you) can nudge its future predictions in various directions, right now, by investing in new experiences. You are continually cultivating your past as a means of controlling your future."

For me this lesson on how the brain works and how we can improve our future is the closest thing I have to a New Year's resolution: seek out new experiences and randomize the way I do things in an attempt to give myself a little bit better chance at making better decisions.

And now, onto some favorite and popular sections of 2023's SITALWeeks:

One of the more popular reads was last year's kickoff post in #379 that discussed the differences between innovation in the purely digital world versus the analog world titled: When Positive and Negative Feedback Loops Collide (couple this read with Zippy Drones from #389).

Also from #379 was a post on the derivative impacts of GLP-1s titled: The Impact of Eating Less on Food Supply Chains and Healthcare. Later on, in #412's Diet Drugs Dampening Snack Spend we found out GLP-1s are already starting to measurably reduce caloric consumption.

In #380, the popular post Follow the Developers - in order to find the next winning platforms - took on increased meaning as the year progressed and LLMs entered the mainstream. 

In More Q, Less A in #382, I dove into the challenging art of learning how to ask better questions. This exploration of Beginner's Mind, Socratic Questioning, and Rhetoric is something that is still very top of mind for me today as we navigate a rapidly evolving and unpredictable future. 

One of my favorite posts was You Auto-Complete Me in #385 that explored the nature of human consciousness as illuminated by LLMs. Couple that with AI Awareness from #387 that explores the relationship between humans, homeostasis, and AI. Also on this thread, I looked at the twisted Comedic Brain and AI's sense of humor in #396. Many of these posts cover my sense that AI embodied in robots will be the most important development in the coming decade. 

I looked at the evolution of machine interfaces from multi-touch to chatbots in #391's Discovery Engines. Couple this with #394's Passing the Innovation Baton. This shifting user interface and changing gateways is something I think of often. See also: Smart Phones, Dumb Pipes.

I contemplated the future of investing in Vanishing Edges as the nature of markets evolves. 

David Bowie, Bob Dylan, and Buster Poindexter taught me about the nature of reality in Cinéma Vérité.

Long time readers know that I am obsessed with the evolving nature of content and our increasingly fragmented relationship with it. I explored this idea in depth in "Will it Play in Peoria?" in #403.

I discussed the current AI bubble in Lessons from an EDFA in #404, and I also discussed the Magnificent Seven's tenuous dominance of financial markets in #418. 

I covered a farewell to one of the greatest performance artists of all time in “I Know You Are, But What Am I?”.

Another fan favorite post was #409's The Simulacrum, which explored the idea of using AI to slow down time as the number of interacting agents in the economy explodes. Couple this with Jobless Economic Growth? in #420.

Artificial Efficiency in #414 looked at the impact of AI on headcount and productivity, something I suspect will be a big topic for the next several years. 

I really enjoyed writing about what we can learn from the old TV ads of my youth in Lessons From Vintage Advertising.

Later in the year, I began exploring the death of journalism and what it means for society as the large Internet gatekeepers all pushed to demonetize news in No News Is Bad News.

Finally, a recurring theme last year was my ongoing discussion of how AI is likely to create a new Age of Wonder for scientific discovery. See also: Science GPT.

Happy New Year!

✌️-Brad

Disclaimers:

The content of this newsletter is my personal opinion as of the date published and is subject to change without notice and may not reflect the opinion of NZS Capital, LLC.  This newsletter is an informal gathering of topics I’ve recently read and thought about. I will sometimes state things in the newsletter that contradict my own views in order to provoke debate. Often I try to make jokes, and they aren’t very funny – sorry. 

I may include links to third-party websites as a convenience, and the inclusion of such links does not imply any endorsement, approval, investigation, verification or monitoring by NZS Capital, LLC. If you choose to visit the linked sites, you do so at your own risk, and you will be subject to such sites' terms of use and privacy policies, over which NZS Capital, LLC has no control. In no event will NZS Capital, LLC be responsible for any information or content within the linked sites or your use of the linked sites.

Nothing in this newsletter should be construed as investment advice. The information contained herein is only as current as of the date indicated and may be superseded by subsequent market events or for other reasons. There is no guarantee that the information supplied is accurate, complete, or timely. Past performance is not a guarantee of future results. 

Investing involves risk, including the possible loss of principal and fluctuation of value. Nothing contained in this newsletter is an offer to sell or solicit any investment services or securities. Initial Public Offerings (IPOs) are highly speculative investments and may be subject to lower liquidity and greater volatility. Special risks associated with IPOs include limited operating history, unseasoned trading, high turnover and non-repeatable performance.

SITALWeek #420

Welcome to Stuff I Thought About Last Week, a personal collection of topics on tech, innovation, science, the digital economic transition, the finance industry, and whatever else made me think last week.

Click HERE to SIGN UP for SITALWeek’s Sunday Email.

In today’s post: regulators are chipping away at Android and iOS, resulting in a steady opening up of the platforms; drone defibrillators to the rescue; puzzling content gains steam with kids on YouTube; Meta goes all in on RISC-V; AI bites the Internet hand that feeds it; rise in US pedestrian deaths at dusk; investors can't beat the M7; falling rents portend a Fed that is behind the curve; what will an economy without job creation look like? And, much more below...

SITALWeek will be on holiday break with the next new issue posting on Sunday January 7th. Happy holidays to all my readers!
 

Stuff about Innovation and Technology
Avoid High-Rise Heart Attacks
An 11-month experiment using drones to deliver automated external defibrillators (AEDs) to patients with cardiac symptoms has now been published. I wrote about the trial while in progress last year, and the overall results show that AEDs could arrive around three minutes faster than emergency personnel in the majority of cases and, in one instance, saved a patient's life. That margin is significant because, in a cardiac arrest situation, every minute that passes decreases survival by around 10%. Researchers suggest that a “semi-urban” population center is the target environment for drone usage (due to a sufficient number of calls and slower ambulance response times). They also note several drawbacks to using drone-deployed AEDs, such as bad weather, a reliance on untrained bystanders, and calls from high-rise buildings (which the drones couldn’t access). Fun fact: it turns out that the higher you live in a building the lower your survival rate of cardiac arrest, and, in one study, no one above floor 25 survived.
 
Baby Shark Gets Flushed
The Baby Shark video producer CoComelon is missing its forecasted revenue goals to such an extent that the debt-saddled parent company, Candle Media (helmed by the two former Disney deputies who are now advising Bob Iger on a Disney turnaround), is seeking to refinance its liabilities. While other factors are likely contributing to Candle Media’s woes, including the prolonged Hollywood strikes earlier this year, the market for premium productions, as we know, is being washed out and devalued by the infinite content of our time. The Washington Post gives us a prime example of this floodtide: Skibidi Toilet. Ever heard of it? Me neither. But, apparently, its videos have been viewed 65B times this year, making it a top trend on YouTube with a rapidly growing base of 36M subscribers. As the WaPo explains: “The series tells the story of Skibidi Toilets (toilets with human heads) engaged in a war with people who have CCTV cameras, speakers and televisions for heads amid a dark and dystopian landscape. They battle each other across an expanding industrial world that includes New York City landmarks, with the Skibidi Toilets acting on behest of their leader, G-Man, to destroy humanity and transform more people into Skibidi Toilets.” Still not clear on what it is? Watch this recent episode that garnered over 6M views in its first ten hours...if you dare.
 
Smart Phone, Dumb Pipes 
A number of developments on the regulatory front continue to indicate an opening up of iOS and Android. I am reminded of the Telecom Act of 1996, which had a goal of letting “anyone enter any communications business – to let any communications business compete in any market against any other.” While I am not arguing that regulation is going to make iPhones and Android devices the “dumb pipes” of the telecom companies, I am suggesting that the walled garden and vertically integrated nature of the platforms could be eroding. The NYT walks through the various legal cases Google is facing, including its recent loss to Epic. Epic, maker of the game Fortnite, has been challenging Apple and Google over monopoly control of app stores (Epic is still appealing a case ruled in Apple’s favor). Apple also recently agreed to open up NFC capabilities on iPhones to rivals in Europe. What might really empower rivals, however, is access to the proprietary AI chips and processing on smartphones, which could create an entirely new innovation cycle for app developers that vastly increases the value of smartphones. If Apple and Google are, however, able to maintain the status quo by keeping phone AI processors to themselves (or limit developer access to it), it would preserve a significant barrier to competition. For example, Google’s Gemini Nano AI is available today exclusively to Google apps on Google’s Pixel 8 Pro handset; but, I am sure developers would love to have native access to it for their apps. Or, consider Microsoft’s new small language model Phi-2, which is said to outperform Gemini Nano; however, will Microsoft be able to access the specialized chips that would enable the SLM to run on Android or iOS devices? Limiting access to AI processing on mobile devices could force Microsoft, Meta, and others to pursue creation of their own smartphones. While Apple and Google might have something to lose by opening up their mobile platforms (Apple far more so than Google given the latter largely derives its revenue from search and YouTube), they could have far more to gain by opening their platforms and lowering the tariffs they charge for access. As phones are likely to serve as hubs for more connected devices like AR glasses, open access should enable even more innovation for the next waves of technology. 
 
Will Generative Search Sideline its Teacher?
The shift to generative-AI-powered Google search results could cause a decline in traffic to news publishers by around a third. Google overall accounts for 40% of traffic to news sites, and an assessment indicates this figure could be cut by 75%, leaving Google as a minor player in facilitating only 10% of the news traffic to publishers. I’ve been using Google’s Search Generative Experience (SGE) beta since it launched months ago, and it has greatly reduced how frequently I click through to actual web pages. This search shift is exacerbating the trending reduction in traffic to legitimate news sources, which has already been responsible for upending many formerly profitable business models (see No News is Bad News). There is a Catch-22 here: AI models are smart because they were largely trained on web content, so if the web content creation diminishes over time because of AI use, can AI still be smart and useful? Meanwhile, BI reports on the rise in people who read news from legitimate sources to their followers on TikTok. But, this trend might be short lived given the rise of AI newsrooms. Just try to watch this video of an entirely AI newscasting technology without imagining just how easy it will be to fabricate news and make it seem legitimate and true. The AI news engine, Channel 1, is launching in March and plans to augment AI news anchor personalities with LLMs and memories, according to Deadline. They ultimately want to produce custom AI news feeds for each user (including multiple language formats), enabling each of us to end up in our own little bubble of reality, untethered from any shared culture or truth. Today, Channel 1 isn’t much beyond an advanced video game engine, but the concept of creating unique personalities powered by LLMs that can generate the news would be a leap forward in AI content that should make your local news personalities anxious.
 
One RISC-V Basket
Meta has shifted their chip design roadmap entirely to open-source RISC-V, opting to leapfrog GPUs and CPUs even for applications like AI model training and inference. This is a notable recognition of a significant advancement in the RISC-V progression. There are reasons to still be skeptical, as The Next Platform notes, particularly since RISC-V falls short on matrix math, which is key for AI. Meta’s Senior Director of Engineering, Prahlad Venkatapuram, appears confident the hurdles can all be overcome for RISC-V AI chips.

Miscellaneous Stuff
Nocturnal Pedestrians Beware
After decades of decline, pedestrian deaths caused by cars have risen steadily (and uniquely) in the US since 2009. The increase is largely driven by incidents that occur after dark, according to the NYT. Nighttime pedestrian death rates are flat or down in other countries over the same period (2009 to 2022) that the US has seen an increase from 9 to 16 deaths per million people. What’s different in the US? While there are many theories, none are conclusive. The timing in the change is of course coincident with the invention of smartphones, but this too is true of other countries. But, perhaps transmissions play a role: only 1% of cars in the US are manual vs. 70-75% in Europe, and it’s harder to futz with your phone while driving a stick. The evening commute also appears to coincide with the time of day when people spend the most time on their phones while driving. Surprisingly, despite our apparent passion in the US for driving monster-truck-sized vehicles everywhere, vehicle size does not appear to be a factor. There is some evidence that the Southern migration in the US has skewed the statistics, as Southern states have suffered the largest increase in deaths, possibly because many sunbelt cities were developed in a post-automobile world and are thus less pedestrian-friendly than Northern cities.

Stuff About Demographics, the Economy, and Investing
M7 Spoil Active Management Returns
According to CityWire, only 9% of global growth equity portfolio managers are ahead of their benchmarks over the last three years. This rolling, three-year stinkfest is down from a recent high in December 2019 of 39%. The biggest factor in the underperformance is the power law of the Magnificent 7. (Most funds are subject to diversification requirements that in many cases prevented them from allocating at or above index weights for the M7). For more on this economic power-law phenomenon, see the final section of #418.

Rent Rate Reductions 
Real-time data from Redfin show that rents in the US fell 2.1% y/y in November. The drop, which comes amidst an increase in construction of new apartments, is notable because rental rates (rent and owner’s equivalent rent) account for around 40% of core CPI, which is a critical component of Fed rate policy. Unfortunately, this reversal in rent trajectories won’t be reflected in CPI data for quite some time, as lagging data can be slow to move due to a high percentage of multi-month/year lease terms. Thus, the Fed is relying on a number that doesn’t reflect the current state of the economy, causing them to react too late when setting rates. Further, the number of newly completed apartments coming to the market and the number of new apartments under construction stand at a three-decade high, implying that further downward pressure on rents is likely. The recent signal from the Fed that it may cut rates three times in 2024 could be too little too late to fend off deflationary trends (or, at the very least, will continue the punitively high rates far longer than necessary). 
 
Jobless Economic Growth?
Last week, the CEO of Walmart, Doug McMillon, noted in a CNBC interview that generative AI is likely to impact job growth at Walmart. (Only a few clips of the interview are online, but I was able to watch the entire interview on CNBC when it aired to understand the broader context of what he said). With respect to stores, McMillon said to expect: “fewer associates at the backend of the store, and eventually the front end of the store, but the sales floor probably gets more associates over time.” And, as it relates to the supply chain, “we see having the same number of people, but their productivity will be much higher.” Growth without headcount increases would be a notable divergence from the historical trend.
 
Ever since I wrote about the rise in revenue per employee (and revenue growth without meaningful hiring) at major AI companies like Microsoft and Google, I’ve been thinking about what it means to have economic growth without job growth. While productivity and other economic forces can lift everyone’s boats (especially corporate profit margins and the stock market), ultimately, we should want more people participating in the economy to drive growth. Can AI that displaces jobs be an engine of economic growth, or will slower job growth (I am not so pessimistic as to think that AI will cause job opportunities to decline in the long run, but it’s certainly in the range of outcomes) offset some of the productivity boost to the overall economy? The short answer is the same as it always is with respect to forecasting the complex adaptive system that is the economy: nobody knows! But, thinking about the question seems worthwhile. 
 
Reductively simplifying the Walmart example, what does it mean for Walmart to grow revenues if they aren’t creating jobs for people who shop at Walmarts? Perhaps the shift will be toward profit growth (more so than revenue growth) and stock market increases, propelling the economy while also further splitting the haves and have-nots. Perhaps a shift to a distributive economy, rather than a growing economy, is around the corner as a result of jobless economic growth. With more people retiring than entering the labor force in the near future due to declining birth rates, economic growth without job growth seems existential to the economy. Still seeking answers, I decided to query Bard, now powered by the new Gemini Plus. Bard’s answer to “how can an economy grow without job growth?” contains several interesting and germane points regarding our current economic situation:
 
While an economy can theoretically grow without job creation, it is a very limited and unsustainable scenario. There are a few specific situations where this could occur, each with its own limitations:
1. Automation and Productivity Growth:

• Technological advancements can automate tasks previously done by humans, leading to increased productivity without increasing employment. This can temporarily boost economic growth through increased output and efficiency.

• However, this often leads to long-term unemployment and income inequality, as displaced workers struggle to find new jobs in the changing economy. This can ultimately stifle economic growth due to reduced demand and social unrest.

2. Resource-Driven Growth:

• Economies heavily reliant on natural resources like oil or minerals can experience growth driven by the extraction and sale of these resources. This can create pockets of wealth and development, but often doesn't translate into widespread job creation or economic diversification.

• When resource reserves deplete, the economy can suffer a major downturn without the diversification needed for a smooth transition.

3. Debt-Fueled Consumption:

• An economy can artificially inflate growth by borrowing heavily and using the funds to increase consumption and investment. This can create a temporary boom, but it is unsustainable in the long run.

• Eventually, the debt burden becomes too large, leading to financial instability, economic crisis, and austerity measures that can destroy jobs and shrink the economy.

4. Rent-Seeking and Unequal Distribution of Wealth:

• An economy can grow when a small group of individuals or businesses capture a disproportionate share of the wealth through rent-seeking activities like monopolies or special privileges. This can lead to increased wealth concentration without significant job creation or income growth for the majority.

• This inequality can lead to social unrest and political instability, ultimately harming economic growth and stability.

It is important to note that these are just theoretical scenarios and have rarely occurred in real economies for any significant period. In the long run, a healthy and sustainable economy needs both economic growth and job creation to thrive. This requires policies that promote innovation, skills development, entrepreneurship, and equitable distribution of the benefits of economic growth.
Here are some additional factors to consider:

• The role of government spending: Government spending on infrastructure, education, and research can stimulate economic growth and create jobs.

• The importance of international trade: Openness to international trade can promote economic growth by creating new markets for goods and services.

• The impact of monetary policy: Interest rates set by central banks can influence economic activity and job creation.

• The need for social safety nets: Strong social safety nets can help individuals and families cope with unemployment and economic hardship.

In conclusion, while an economy can experience temporary growth without job creation, it is not a sustainable or desirable outcome. For long-term economic prosperity and social stability, economic growth needs to be accompanied by job creation and equitable distribution of wealth.
 
I continue to think a step-function increase in the pace of innovation and technological progress – driven by humans working with AI bots – is likely to create material economic growth and new opportunities that benefit everyone in the economy. This would parallel and surpass the Romantic age of scientific progress that I wrote about in AI Age of Wonder: 
...the Romantic age of science in the 19th century – the second scientific revolution that was inspired by artistic Romanticism. The Romantic movement was a period marked by an interest in and an awe for the natural world. The 1800s saw Darwin’s theory of natural selection, the discovery of electromagnetism, germ theory, the telephone, etc. Scientists were heroes, but also sometimes seen as pressing boundaries too far – Mary Shelley’s 1818 character of Dr. Frankenstein is the most obvious such example. This so-called Age of Wonder saw a vast amount of scientific discovery that also entered into and impacted the common culture. While (from our current standpoint) it was a relatively slow progression of scientific discovery and communication, it likely seemed fast paced to those caught in its wake, with discoveries having dramatic impacts on society. If I contrast this period with the world today, it seems like we could have an AI Age of Wonder several orders of magnitude greater in both the number and speed of discoveries. This increased amplitude and compressed timeframe is likely to have extremely unpredictable outcomes.
As I wrote about in more detail in Simulacrum, ultimately, an agent- or bot-based economy driven by autonomous decisions and services could amplify the marketplace from billions to trillions and spin off value for the real-world economy in which we all exist.

✌️-Brad

Disclaimers:

The content of this newsletter is my personal opinion as of the date published and is subject to change without notice and may not reflect the opinion of NZS Capital, LLC.  This newsletter is an informal gathering of topics I’ve recently read and thought about. I will sometimes state things in the newsletter that contradict my own views in order to provoke debate. Often I try to make jokes, and they aren’t very funny – sorry. 

I may include links to third-party websites as a convenience, and the inclusion of such links does not imply any endorsement, approval, investigation, verification or monitoring by NZS Capital, LLC. If you choose to visit the linked sites, you do so at your own risk, and you will be subject to such sites' terms of use and privacy policies, over which NZS Capital, LLC has no control. In no event will NZS Capital, LLC be responsible for any information or content within the linked sites or your use of the linked sites.

Nothing in this newsletter should be construed as investment advice. The information contained herein is only as current as of the date indicated and may be superseded by subsequent market events or for other reasons. There is no guarantee that the information supplied is accurate, complete, or timely. Past performance is not a guarantee of future results. 

Investing involves risk, including the possible loss of principal and fluctuation of value. Nothing contained in this newsletter is an offer to sell or solicit any investment services or securities. Initial Public Offerings (IPOs) are highly speculative investments and may be subject to lower liquidity and greater volatility. Special risks associated with IPOs include limited operating history, unseasoned trading, high turnover and non-repeatable performance.

SITALWeek #419

Welcome to Stuff I Thought About Last Week, a personal collection of topics on tech, innovation, science, the digital economic transition, the finance industry, and whatever else made me think last week.

Click HERE to SIGN UP for SITALWeek’s Sunday Email.

In today’s post: KISS goes virtual; Google's Gemini hints at a plateau in LLMs, but moving to a new level could happen with the integration of AI and robotics; Mark Cuban's is exerting pressure on big pharmacy; melatonin use soars; Juror #6 gets a production deal; military goes high tech. And, much more below...

Stuff about Innovation and Technology
Cloning KISS
The legendary rock band KISS played their last concert and transcended into the virtual world – to live on as full-body motion capture 3D virtual characters. The new virtual KISS was created by Disney’s Industrial Light & Magic. The band sat down for a video conversation with ILM execs to discuss their history and decision to immortalize themselves. The band has not indicated what the virtual group will do in the future, and it’s not clear to me who owns the band’s likeness given the mocap was funded by Sweden-based Pophouse Entertainment. While the technology used by ILM to digitally clone the band is run-of-the-mill superhero movie stuff, the idea that the band is retiring from public concerts and handing the reins to a licensed, virtual version of themselves, while still alive, could set a precedent that other performers may follow. We are already drowning in infinite content, and the proliferation of virtual versions of real performers could cause the flood waters to rise even more. 
 
Gemini’s Plateau
Google finally unveiled their new multimodal AI called Gemini. A regular “Pro” version has already been integrated into Bard, and the “Ultra” version, dubbed Bard Advanced, will be available next year. In my experience, the new Bard is a huge upgrade from prior versions and is remarkably faster than ChatGPT. I would rate its capability as somewhere between GPT3.5 and GPT4, but more like a faster version of 3.5. However, the real magic appears to be in the Ultra model as seen in this video (the video is highly modified for effect, but I think reading through the process makes the AI seem even more impressive). The stats behind Gemini Ultra indicate it surpassed prior models on 30 out of 32 top AI benchmarks, and it also scored 90% on the Massive Multitask Language Understanding test, besting human experts across a variety of language, science, and social topics, such as ethics. The model also outperformed 85% of software programmers in coding competitions (although competitions differ from actual coding work). Gemini is a combination of an LLM and DeepMind’s tech that powers AlphaGo and AlphaFold. “Multimodal” means it works with text, images, video, sound, code, etc. I would guess that Gemini Ultra slightly exceeds the capabilities of the current GPT4 Turbo, but its multimodal processing appears stronger. 
 
While the base-case expectation is for LLMs to become superhuman, it seems like it might be harder to progress much beyond regular human skills, simply because today’s LLMs are so human-like. Afterall, the best of us humans still make countless mistakes, so we wouldn’t want to rely on a sole LLM for anything necessitating a high degree of accuracy. Several companies have been able to achieve GPT3.5-level functionality, but only two have had the ability to get to the GPT4 Turbo/Gemini Ultra level of performance. It’s plausible models will plateau at this point without some additional breakthrough, leaving us with advanced – but not superhuman – AI. And, I doubt we would want to rely on this level of AI to perform critical tasks, e.g., even OpenAI admitted last week they aren’t sure why GPT4 has started acting “lazy”, but are looking into it. 
 
Perhaps an important step forward will be going beyond sight and sound by integrating various sensors and data feeds to create an intelligence with super-sensing capabilities (e.g., LIDAR, weather forecasts, vibration, etc.). Gemini has an on-device version called Nano that will soon launch on Pixel 8 phones. On my Pixel 8, I’ve been noticing a lot of helpful copilot behavior already, such as real-time summaries of web pages and the ability to proofread and edit what I type with the click of a button in the Google keyboard. As I often note, the winner in any new platform shift is the one that wins developers and users, with developers who are creating apps for users being the key target. Microsoft OpenAI has an early lead, but Gemini Ultra could pull ahead, especially based on its apparent speed and advanced multimodal processing. This is a real-time arms race unlike anything we’ve seen before (e.g., Android came years after iOS, and Apple has still powerlawed app-economy profits). In a brief interview with the Alphabet and DeepMind CEOs, there seems to be an implication that the next version of Gemini (currently being trained) will be a significant leap from what we’re seeing in the demo video. That speed of progress is hard to wrap your head around – if AI does not soon plateau, then this is the most nonlinear innovation cycle I can recall in the history of the tech sector. 
 
Seeing Gemini has me thinking about my post on embodying AI in the physical world from March:
 
Much of being human involves processing myriad inputs from our expanded seven senses (sight, hearing, smell, taste, touch, thoughts, and emotions, all of which feed into creating our ongoing sense of self). But, ChatGPT runs on a server in some Microsoft data center. It’s effectively got one sense – the interaction between the model it was trained on and the input from a human interlocutor. We could describe Bing-Chat as having two senses, with the second being its ability to access the Internet in real time. We could theoretically feed more sensory data into ChatGPT, in particular real-time images/video, sounds, or even things that approximate “touch” like temperature and pressure data. Or, we could embody the LLM in some type of physical form that allows it to more dynamically interact with, and receive input from, the real world. It’s an open question as to whether LLMs would learn to process and respond to data in a human-mimetic way or if more alien behavior would emerge.
 
Neuroscientist Antonio Damasio characterizes human existence in terms of a drive toward homeostasis. He sees this quest for comfort as the fundamental life force (detailed in his book The Strange Order of Things). Essentially, the nervous system is a connected tool to make the living organism (e.g., a human) feel in balance. If we need calories (or detect that free calories are available), we feel hungry and eat. If we are cold, we seek shelter. Damasio further believes feelings emanate from the drive toward homeostasis and are a way for the brain to interpret good or bad states and act on them. He speculates we can derive all of human consciousness and culture from our thoughts, feelings, and actions regarding our relationship with homeostasis. I tend to agree with him, but I also hold these beliefs loosely given that we don’t know all the answers. Thus, homeostasis seems intertwined with feelings, consciousness, and a physical body able to monitor (and react to) our internal state and the world around us. Therefore, the critical junction of combining a LLM with a physical form capable of monitoring its systemic and real-world inputs (e.g., temperature, pressure, proprioception, energy reserves) – and react to these sensory data in a way that seeks to maintain its own, human-equivalent homeostatic targets – seems like the logical next step in the trajectory of AI. And, we may already be rapidly progressing down that road. Microsoft’s Autonomous Systems and Robotics division announced their intentions to put OpenAI technology into robots (blog post and video). This integration could ultimately lead to a paradigm shift in AI where we go from programming a specialized robot/tool to do a specific task with specific inputs, such as autonomous driving, to a situation where you could just ask ChatGPT to go learn how to drive a car. However, merging AI with a physical form demands an exceedingly careful approach for successfully progressing such tools in the real world, particularly with respect to protecting human safety (a couple weeks ago, I shuddered at the idea of integrating ChatGPT into the new Boston Dynamics humanoid!). Rather than a thoughtful approach, we unfortunately appear to be heading toward what I described last week as “my AI can beat up your AI”, with tech leaders now attempting to create rival AIs backed by distinct ideologies. Bill Gates is apparently heavily involved in leading the strategy as an advisor to what he refers to as “Microsoft OpenAI” in this FT podcast interview (the transcript is here, but it has several typos). Gates largely dismisses concerns over AI in the podcast. Given Microsoft’s desire to embody LLMs in real-world robots, this blasé stance is concerning to me. Gates went so far as to question whether we should blame people, rather than the AI itself, for its shortcomings. While this mentality also concerns me, the underlying point – that the risk with AI is more weighted to how people will use it than the AI itself – is valid. Gates described the pending GPT4 as: “wow”, with capabilities coming “many years before I expected”.
 
The next-level scenario of embodied, human-mimetic chatbots brings emergence to mind. Emergent behavior is something new that occurs in a complex system of interacting agents that wouldn’t have happened (or been predicted) based on the agents’ isolated actions. Certainly, chatbots are an emergent phenomenon from LLMs, but I am not sure today’s chatbots themselves demonstrate emergence (although I will note that the Microsoft researcher in that video linked above declared that they do, but I do not know what definition he was using). People are certainly finding emergent use cases for chatbots. When LLMs become embodied in the real world, however, we should expect to see emergent phenomena from the robots themselves. For this reason in particular, the scarcity of caution today is concerning. 
 
CVS’s Cost Concessions
The success of Mark Cuban’s transparent pharmacy company is putting broader pressure on the prescription-peddling industry. Drugstore giant CVS announced last week they are adopting a new prescription pricing strategy more aligned with that of the Mark Cuban Cost Plus Drug Company. This move comes after one of CVS’s customers, Blue Shield of California, migrated to Cost Plus. CVS is also likely responding to FTC scrutiny of its pharmacy benefit management (PBM) business, which, as far as I can tell, is designed to obfuscate pricing info and cause drugs to be more expensive than they need to be. We first covered the Mark Cuban Cost Plus Drug Company nearly two years ago, and it’s great to see the positive progress and impact it’s having. The company intentionally set out to work around the clouded and misaligned PBMs and insurance company agendas. Even with the reactive pricing changes, it seems unlikely that a business like CVS, with all of its conflicts, can best a highly non-zero-sum-focused business like Cuban’s. As Business Insider writes: “It's not clear if CVS's new program will result in lower prices overall for consumers. Some drugs will cost more under CostVantage, while others will cost less, according to the Wall Street Journal. The costs could also depend on the terms of an individual's insurance plan, and whether they choose to use it.” As for Cuban, he is reportedly selling a stake in the Dallas Mavericks (although will continue to run the organization) and announced he will be exiting the hit TV show Shark Tank. I am looking forward to seeing what he does next.

Miscellaneous Stuff
Epidemic Insomnia
According to research published in JAMA Pediatrics, 19% of US children aged 1 to 14 are relying on melatonin to aid in sleep, up from 1.3% 5-6 years ago. The nearly 15x increase seemed to ramp during the pandemic. One suggestion is increased screen time (which can disrupt the body’s natural melatonin cycle) is to blame, or perhaps it’s a general increase in anxiety (or maybe it’s TikTok, which seems able to increase both screen time and anxiety). Melatonin is available over the counter in the US, but other countries, like the UK, require a prescription. A total of 27% of US adults are now said to take melatonin.
 
Gladden’s Good Karma
Ronald Gladden, the earnest and unwitting star of the reality/mockumentary TV series Jury Duty, has signed a two-year development deal with Amazon MGM for new projects. From #396:
Jury Duty is a new bingeable show on Amazon’s Freevee channel (I watched it via Prime Video). The show follows a fake jury trial where only one member of the jury, Ronald Gladden, isn’t in on the act. Gladden, a 29-year-old solar installer, spent weeks in fake sequestration under the assumption he was part of a documentary on the judicial process. It’s a bit like a real-world version of The Truman Show, and it has many delightful moments throughout. I of course loved it because it plays with society’s increasingly blurred lines between fact and fiction (see Cinéma Vérité). In yet another reality-defying twist, Amazon plans to submit Ronald Gladden for a best-actor-in-a-comedy-series Emmy, despite him not even knowing he was in a TV show during filming. (Note: Jury Duty received four Emmy nominations, but not for best actor).

Stuff About Demographics, the Economy, and Investing
AI Called to Duty
Anduril’s new Roadrunner and Roadrunner-M jet-powered autonomous drones are designed to destroy other jet-powered targets, like enemy drones and missiles, mid-flight. The reusable vehicles (which launch like a rocket and then return home after completing their mission) are one potential advance beyond the US’ analog-focused and outdated military efforts. Development of this type of intelligent drone aligns with what I wrote a couple months ago: 
Autonomous drones imbued with AI are becoming quite common, as we’ve discussed in the past. The WSJ reports that the Pentagon wants to build a fleet of thousands of AI robots for air, land, and sea deployment. The move is said to counter China, which is far ahead of the US with these capabilities. This is starting to feel like the new “mutually assured destruction”, i.e., every country will have a massive fleet of AI-military tech, and the consequences of anyone deploying it would be met with just as big of a threat.
The main difference of course is that much of the autonomous and AI-driven technology is low cost and available to anyone with minimal resources. That is a very different situation than a handful of nations controlling the supply chains and production for advanced military weapons and nuclear bombs.

✌️-Brad

Disclaimers:

The content of this newsletter is my personal opinion as of the date published and is subject to change without notice and may not reflect the opinion of NZS Capital, LLC.  This newsletter is an informal gathering of topics I’ve recently read and thought about. I will sometimes state things in the newsletter that contradict my own views in order to provoke debate. Often I try to make jokes, and they aren’t very funny – sorry. 

I may include links to third-party websites as a convenience, and the inclusion of such links does not imply any endorsement, approval, investigation, verification or monitoring by NZS Capital, LLC. If you choose to visit the linked sites, you do so at your own risk, and you will be subject to such sites' terms of use and privacy policies, over which NZS Capital, LLC has no control. In no event will NZS Capital, LLC be responsible for any information or content within the linked sites or your use of the linked sites.

Nothing in this newsletter should be construed as investment advice. The information contained herein is only as current as of the date indicated and may be superseded by subsequent market events or for other reasons. There is no guarantee that the information supplied is accurate, complete, or timely. Past performance is not a guarantee of future results. 

Investing involves risk, including the possible loss of principal and fluctuation of value. Nothing contained in this newsletter is an offer to sell or solicit any investment services or securities. Initial Public Offerings (IPOs) are highly speculative investments and may be subject to lower liquidity and greater volatility. Special risks associated with IPOs include limited operating history, unseasoned trading, high turnover and non-repeatable performance.

SITALWeek #418

Welcome to Stuff I Thought About Last Week, a personal collection of topics on tech, innovation, science, the digital economic transition, the finance industry, and whatever else made me think last week.

Click HERE to SIGN UP for SITALWeek’s Sunday Email.

In today’s post: the virtuous circle of scientific discovery and AI; Hollywood gets in on the ambient content game; Jaron Lanier on compensating the humans that AI feeds on; reflecting on power laws in the stock market this year; Jon opines on semis. And, much more below...

Stuff about Innovation and Technology
ScienceGPT
What the Internet did for accelerating the velocity and transparency of information, AI will do for scientific discovery. The Argonne National Lab’s supercomputer is working on training a trillion-parameter research-focused LLM called AuroraGPT. The compute system, which has over 80,000 Intel CPUs and GPUs combined, will process scientific data, research, and papers with the intent of creating an assistant to speed up basic research. Microsoft is working with the lab to scale up the training to all 10,000 nodes, which could take several months to train to completion. Meanwhile, DeepMind is releasing the models of several-hundred-thousand novel crystalline structures that could revolutionize fields like solar panels, superconductors, batteries, semiconductors, and more. The trove would have taken humans 800 years to discover at our pre-AI pace. The real wild part comes when you start to feed this type of data into something like AuroraGPT to identify potentially viable inventions and breakthroughs for humans to further explore. Researchers at Berkeley have already taken some of the suggestions from DeepMind and used an LLM to suggest synthesis recipes for 58 target materials. Using an automated lab, 41 of the targets were created. The abstract of the paper concludes: “The high success rate demonstrates the effectiveness of artificial-intelligence-driven platforms for autonomous materials discovery and motivates further integration of computations, historical knowledge and robotics.” While I often talk about the long time lapse between a digital breakthrough and its impact on the physical world (see When Positive and Negative Feedback Loops Collide), it’s probably time to start thinking about a faster pace of disruption and innovation affecting a broad swath of traditionally slow-grinding economic gears. The Internet and smartphones created digital advertising, digital media, ecommerce, cloud software and more, but these will likely all pale in comparison to the accelerated scientific revolution coming to the physical world of materials and biology. 

Lanier’s Lens
This brief interview with VR/AI-pioneer Jaron Lanier has a few interesting bits on generative AI. While I am generally in favor of editing, unfortunately the video appears to have been painfully edited to cut away right when Lanier was about to get to the interesting stuff. I think there are some people who should have an uncompromised license to expound, and Lanier is one of them. Around ten minutes in, Lanier talks about the potential to compensate people for contributing to AI models via attribution. The slightly awkward thing here is that Lanier works for Microsoft, which has the potential to use its new copilots to train on people who then will lose their jobs to those same AI copilots. Lanier’s comments could be construed to argue that Microsoft should be paying Windows users to train their copilots, and then giving them royalties in perpetuity. What are the odds of that happening!? And, despite working closely with Sam Altman and OpenAI, Lanier is quick to challenge Altman’s Worldcoin business, which Lanier characterizes as: “Sam wants to do this thing of a universal eye scan based cryptocurrency coin to reward people once AI does all the jobs; obviously that's not within my frame of recommendation. I don't think that's a good idea. I think that some criminal organization will take that over no matter how robust he tries to make it. Like look at crypto, crypto is mathematically perfect and then at the edge it's all criminals and fraud and incompetence.”

Doing More with Less
Last week, I mentioned that Slack had, at one point, tried to subsume the user interface for enterprise software apps, and I voiced my view that chatbots/copilots will all eventually just talk to each other, cutting out the humans in the middle. I came across an interview with the Slack CTO and co-founder that discussed Slack’s plan to integrate more AI models: “So I think almost every kind of information worker role is going to be augmented by AI in some way over the next couple of years. And we're really just starting to scratch the surface. Of course, because these tools are being built by software engineers, software engineering is the first area that's augmented by it. But I think we're going to continue to see it across a wide variety of disciplines of taking some portion of that work, and being able to automate it, make it faster or accurate, and leave time for the higher leverage work.” That last bit about freeing up time sure sounds like a recipe for doing more with fewer employees, similar to what we’ve seen in software engineering’s early adoption of AI copilots. Cynicism about job losses is likely to lose out to optimism long term as AI creates a larger economy with more opportunities, but it’s getting harder to be optimistic about the prospects for many rote, white collar, computer-based jobs over the next few years. (I hesitate to link directly to this interview because the website appears heavily spammy on ads and requires readers to opt into personalization just to see the article, but if you want to track it down it’s with Cal Henderson on a site called Moneycontrol.)

Miscellaneous Stuff
Ambient Hollywood
In #404, I joked that Disney’s Park visitors uploading walking tours to YouTube were creating competition for Disney’s expensive movies and streaming shows. Well, not to be outdone, Hollywood has been getting in on the ambient video business over the past year. On Paramount+, you can chill to three hours of the Star Trek Enterprise NCC-1701-D warp core or watch a crackling fire by a lake in front of Paramount Mountain. Max, meanwhile, is carrying a chill, background music/animation show called [ambient swim]. And, on Disney+, you can now find many Scenescapes Ambiance channels with sounds and backdrops from the Haunted Mansion, Star Wars, Dory’s Reef, and Arendelle Castle. Perhaps these low-budget programs are all preparing us for our inevitable future of 24/7 AI-generated ambient content.

Stuff About Demographics, the Economy, and Investing
M7 Power Law
One of the key tenets of Complexity Investing is that we cannot reliably predict the future; therefore, we instead focus on fundamental characteristics of businesses like adaptability and non-zero sumness to assess investment potential. Likewise, there is very little, if any, useful information to be garnered from current market asset prices, as those numbers generally reflect an array of opinions from countless people, and increasingly AI, all based on chaotic inputs to arrive at a random price that reflects supply and demand in the moment. The best we can do as investors is to rely on as few predictions as possible and a handful of truths from complex adaptive systems to suss out potentially interesting assets. That leaves the final piece of the investing puzzle: when to join those assets on their journey through time and when to leave them, i.e., when to buy and sell. This last bit complicates the picture greatly, of course, and it opens the door for a wide variety of cognitive biases and compounding mistakes that build on each other over time. Unexpected market events occur with marked regularity thanks to the emergent behaviors of complex adaptive systems. But, occasionally, a trend stands out against the normal noise of ebbs and flows. Right now, for example, there are a very small number of very large companies that are the primary drivers of the global markets. This year, the MSCI ACWI, a broad measure of the global markets, is up 16.74% on a total return basis through November 29th (all numbers below are through the same date). In particular, the growth-focused version of the MSCI ACWI is up 27.68%, while the value version is only up 6.19%. That divergence is not atypical, but let’s drill down further. The US has fared better than the global index so far this year, up 20.29% (using the S&P 500 as a proxy). We can use the Russell 1000 Growth Index (R1G) as a proxy for US growth stocks (the index includes approximately 450 larger companies that are assumed to have higher valuations and higher growth levels), and we see that it’s up 36.58%. That is six times the value index and more than twice the overall return of the global markets. To understand why, let’s look at the so-called “Magnificent Seven”: Alphabet, Amazon, Apple, Meta, Microsoft, Nvidia, and Tesla. The M7 are up 101.42% so far this year! These companies, as of November 29th, had a total market value of just under $12T, representing around 40% of the R1G (according to Bloomberg, the market capitalization of the R1G was $29T as of November 29th). To put even more context on the number $12T, the Russell 2000, which represents the next 2,000 largest US companies just behind the top 1,000, has a total value of just $2.72T. Those 2,000 companies are worth less combined than the current value of either Apple or Microsoft.

A little bit of contribution math tells us that the M7 accounted for roughly two-thirds of the R1G performance this year. Similar math emerges for the S&P500. Various reports indicate that this is the narrowest rally ever for a market that has risen more than 15% in a year and this is the most seven stocks have accounted for the market capitalization in history. We expect power laws to be a natural occurrence in complex adaptive systems, but this one is attention grabbing. To have all of the very largest companies all lead with some of the best performance, in some cases for the same reasons, is certainly not something that happens every year. I fear I will disappoint you by not commenting on what it all means or where it goes from here. SITALWeek has covered the salient trends of the analog-to-digital transformation of the economy (including regulatory capture, network effects, power laws, etc.) extensively, and, clearly, much of this outperformance is understandable through that lens. AI, in particular, has the potential to create larger power-law winners, while at the same time increasing the fatness of the tail to the downside risk outcome as well. So, perhaps the extreme market power law of 2023 is explainable from this perspective, or perhaps we should just go with the obvious joke: the AIs are all influencing the market algorithms to drive up the share prices of their creators! One thing we know about technology cycles is that they can be in one of three phases: overspending bubble, bursting bubble, or post-bubble-burst normalization. At the moment, we are somewhere on the tenuous ground between the first and second phase, but these things are unpredictable and defy the odds for long periods of time. 

Semi Summary
Jon recently gave a talk to the CFA Society of Switzerland on the semiconductor value chain and related geopolitics. You can see the presentation here.

✌️-Brad

Disclaimers:

The content of this newsletter is my personal opinion as of the date published and is subject to change without notice and may not reflect the opinion of NZS Capital, LLC.  This newsletter is an informal gathering of topics I’ve recently read and thought about. I will sometimes state things in the newsletter that contradict my own views in order to provoke debate. Often I try to make jokes, and they aren’t very funny – sorry. 

I may include links to third-party websites as a convenience, and the inclusion of such links does not imply any endorsement, approval, investigation, verification or monitoring by NZS Capital, LLC. If you choose to visit the linked sites, you do so at your own risk, and you will be subject to such sites' terms of use and privacy policies, over which NZS Capital, LLC has no control. In no event will NZS Capital, LLC be responsible for any information or content within the linked sites or your use of the linked sites.

Nothing in this newsletter should be construed as investment advice. The information contained herein is only as current as of the date indicated and may be superseded by subsequent market events or for other reasons. There is no guarantee that the information supplied is accurate, complete, or timely. Past performance is not a guarantee of future results. 

Investing involves risk, including the possible loss of principal and fluctuation of value. Nothing contained in this newsletter is an offer to sell or solicit any investment services or securities. Initial Public Offerings (IPOs) are highly speculative investments and may be subject to lower liquidity and greater volatility. Special risks associated with IPOs include limited operating history, unseasoned trading, high turnover and non-repeatable performance.

SITALWeek #417

Welcome to Stuff I Thought About Last Week, a personal collection of topics on tech, innovation, science, the digital economic transition, the finance industry, and whatever else made me think last week.

Click HERE to SIGN UP for SITALWeek’s Sunday Email.

In today’s post: Disney launches more adorable robots with potential commercial usage; new AI music tools are transformative to the art form; social media's shift away from news is leaving journalism in the lurch; copilots are the new UI and API layer for enterprise software; narrating your life; weather forecasting breakthrough; take a ride in Jeff Bezos' head; is consolidation the reason productivity is stalling in the US? And, much more below...

Stuff about Innovation and Technology
Emotive Bots
Disney’s adorable robots that I wrote about in #413 have already shown up in the “Star Wars: Galaxy Edge” theme park experience. Here is a video of the WALL-E-esque bots delighting guests. Disney used a combination of simulation, reinforcement learning, and 3D printing to rapidly produce bots that mimic personalities developed by character artists, going from concept to live, in-park deployment within a year. From the first US president audio-animatronic figures at the 1964 World’s Fair to these compelling bipedal character bots, Disney is likely one of the few top robotics and machine learning shops in the world, advantaged by their unique approach to robots. As one Disney Imagineer roboticist, Morgan Pope, put it: “We have a very different mission compared to conventional roboticists. We’re trying to use electromagnetism to create emotions.” Pope designs humanoid robots that can flip, jump, and swing around in stunt performances at the parks. Disney also has a new robot called Indestructible that rollerblades, somersaults, and does other stunts. I don’t know what Disney’s plans are beyond theme park attractions, but if you add an LLM to these bots, I think they could sell many millions to retail consumers. Disney’s robots also make me wonder if this more diminutively sized form factor could be of greater utility in commercial applications as well, rather than the full adult-sized bipedals most companies are working toward.

Music Is Dead. Long Live Music.
Google’s DeepMind introduced Lyria, a new set of tools to create music. Hum a few bars and you can transform a melody in your head into an instrumental or song. Another new feature called Dream Track allows users to describe a song they want in words and Lyria will generate it, including AI vocals from real singers like Charlie Puth or T-Pain. This music can then be set as the background for YouTube Shorts videos. While the fear of artists being completely replaced by generative AI is real, in the near term, I am looking forward to artists working with these tools to create new genres of music. The interesting business angle here is the musicians who are licensing their voices to the AI engines: is it a deal with the devil, or a leveraging of new technology? It’s pretty easy to envision how these advances will soon be applicable to video: describe a plot in detail, name your actors and director style, identify some music concepts, and you’ll have your own Hollywood studio. Here, longtime readers won't be surprised that I will yet again plug the 2014 movie The Congress for anyone who hasn’t seen it yet. Jeffrey Katzenberg says it will take 90% fewer people to make a big Hollywood animated movie thanks to AI. AI’s impact on the media industry is moving fast. Tom Hanks, commenting on future AI versions of himself, said: “some people are not going to dig it because it’s not really a real human being, and other people simply aren’t going to care because the story is OK. That’s not that different from any type of tool that [has] come into the moviemaking process since sound.”

No News Is Bad News
Recent shifts in user behavior and algorithm changes have caused traffic at news sources from social media to plummet by over 50% y/y and ~80% from 2020. The user behavior changes come from a shift toward social networking (i.e., TikTok) and video streaming that doesn’t emphasize linking to news/sources, while the algorithm change largely reflects Meta and X’s de-emphasis of news across their properties. It’s perhaps too early to declare the death of clickbait journalism, but it’s certainly taken some body blows and could be fatally wounded if algorithms and user behavior continue on their current trajectory. I believe there are other economic factors that will continue to shift users toward a smaller and smaller number of larger platforms, such as the steady increase in privacy changes on iOS and Android along with the pending loss of third-party cookies in Chrome. The logic is that larger platforms have enough first-party data to attract more advertisers at higher rates, while smaller ones will suffer a downward spiral in market share and ad dollars, rendering them unable to support their content. Media layoffs are up six-fold to ~20,000 people this year, according to Axios. Examples abound, such as Jezebel shutting down after sixteen years. 

Clickbait news was bad, but now people may increasingly get their news “reported” by influencers regurgitating distorted versions through their narrow (and often bonkers) lenses without linking to the original researched and edited news sources. Journalists don’t always get it right, but at least their accuracy rate is higher than that of some random, uncredentialed dude on TikTok. The loss in news traffic and monetization due to privacy changes will further cause journalism to focus on paying subscribers. For example, Insider is retreating back to just Business Insider, with co-founder Henry Blodget stepping down as CEO as the company pivots to focus “more emphasis on serving and attracting subscribers than generating traffic through social-media platforms.” The problem with news subscriptions is that it’s a rather niche business with limited revenues and reach, and, critically, it will exclude the broader group of folks used to having access to ad-supported news. This shift is likely to further polarize and fracture any sense of truth and common culture. I believe we need to reach a breaking point – where no one believes anything anymore – to reset culture, but it’s not clear how that endpoint will manifest, or whether we will reboot to something better or worse. In the meantime, expect an increasingly ignorant general population when it comes to current events.

Copilots Have Landed
GPT and LLM copilots built into operating systems are going to rapidly become the new user interface for consumers (see Discovery Engines), as well as enterprise software users. Imagine all-knowing AI that can talk to all of your enterprise apps, including WDAY, SAP, etc. Well, this completely new business UI is already here. Whereas apps have previously been able to speak to each other through APIs (as well as connective tissue linking programs and data created by large middleware layers like MuleSoft), going forward, custom GPT copilots will interact with workers and each other, creating a richer connection between apps and data. The apps themselves will increasingly retreat into the background as the UI changes from the app itself to a chatbot/copilot, which will shift proprietary value from software interfaces/functionality to exclusively data. Slack attempted such a transition by allowing queries of third-party apps from the console, but, with Microsoft’s Copilot Studio, I think we could see a new era of UI changes. While legacy cloud software will undoubtedly be around for a long time (just like legacy client/server software), its value is likely to be increasingly capped or transformed by AI. Further, once copilots/chatbots get the hang of things, humans will become marginalized in many of today’s popular business processes and workflows. If today’s large enterprise software companies cannot transition from user-based to value-based pricing and leverage proprietary pools of data, they face an uncertain future. 

SLMs
Microsoft has had success creating small language models. In contrast to LLMs, which are trained on a large set of real-world data and history, SLMs, like Microsoft’s Orca, are trained on tailored synthetic data. These lighter weight and lower power models could be deployed in a distributed manner, adding intelligence to a variety of things, like a butter-spreading bot. 

Querying Videos
Google’s Bard can now answer deeper questions about YouTube video content. For example: “if you’re looking for videos on how to make olive oil cake, you can now also ask how many eggs the recipe in the first video requires.” Training GPTs on videos will be key to giving them broader context and preparing them for real-world activities once they are embedded in robots. 

Synthetic Narration
A programmer used GPT-4V’s image processing capabilities and an AI voice simulator to create a live narration of screen grabs of his desktop as he worked, voiced by a synthetic David Attenborough. Add this to something like Meta’s Ray-Ban Smart Glasses and you could have your whole life narrated as you move through the world. Attenborough is none too excited about this development: "it is of the greatest concern to me that one day, and that day may now be very close, someone is going to use AI to deceive others into believing that I am saying things contrary to my beliefs or that misrepresent the wider concerns I have spent a lifetime trying to explain and promote." Luckily for Attenborough, in the not too distant future no one will believe anything is real.

WeatherAI
DeepMind beat the world’s leading weather forecasting model in 90% of cases using less energy than the supercomputers. Previously, I’ve joked that AI weather forecasting might use so much power that it would impact the climate itself, but this breakthrough proves otherwise. “We introduce GraphCast, a state-of-the-art AI model able to make medium-range weather forecasts with unprecedented accuracy. GraphCast predicts weather conditions up to 10 days in advance more accurately and much faster than the industry gold-standard weather simulation system – the High Resolution Forecast (HRES), produced by the European Centre for Medium-Range Weather Forecasts (ECMWF). GraphCast can also offer earlier warnings of extreme weather events. It can predict the tracks of cyclones with great accuracy further into the future, identifies atmospheric rivers associated with flood risk, and predicts the onset of extreme temperatures. This ability has the potential to save lives through greater preparedness.” Google is open sourcing the model.

Miscellaneous Stuff
Absurd Art
Artist Bobby Fingers makes a rowboat out of a giant Jeff Bezos head model. 

Boring Branding
Blockbuster drug names, like Zepbound, are weird in large part because they aren’t supposed to be interesting. 

Telescopic Timeline
The NYT has a great scroll-through of images from the JWST and a brief history of how our imaging of space-based objects has improved over the years.

Stuff About Demographics, the Economy, and Investing
M&A, PE Stifle Productivity and Innovation
Manufacturing productivity has been declining in the US for over a decade. Other Western countries, like Canada, the EU, and Britain, however, are all seeing rising productivity, albeit down materially from the earlier part of the 21st century, according to The Economist. After accounting for all sorts of potential explanations, The Economist mentions a few more speculative guesses. One that caught my attention was the lax antitrust M&A enforcement in the US. It sure seems like there is less competition in a lot of industries, and, in particular, PE-driven M&A has especially shifted corporate strategy to raising prices rather than growing profits through efficiency and innovation (see also The Circle of Debt). This seems to have created a complacency and lack of urgency in many industries to adopt or develop new technologies. That said, we appear to be on the cusp of a major productivity gain in manufacturing from AI-enabled robots, so this trend could reverse in spite of its mysterious sources.

✌️-Brad

Disclaimers:

The content of this newsletter is my personal opinion as of the date published and is subject to change without notice and may not reflect the opinion of NZS Capital, LLC.  This newsletter is an informal gathering of topics I’ve recently read and thought about. I will sometimes state things in the newsletter that contradict my own views in order to provoke debate. Often I try to make jokes, and they aren’t very funny – sorry. 

I may include links to third-party websites as a convenience, and the inclusion of such links does not imply any endorsement, approval, investigation, verification or monitoring by NZS Capital, LLC. If you choose to visit the linked sites, you do so at your own risk, and you will be subject to such sites' terms of use and privacy policies, over which NZS Capital, LLC has no control. In no event will NZS Capital, LLC be responsible for any information or content within the linked sites or your use of the linked sites.

Nothing in this newsletter should be construed as investment advice. The information contained herein is only as current as of the date indicated and may be superseded by subsequent market events or for other reasons. There is no guarantee that the information supplied is accurate, complete, or timely. Past performance is not a guarantee of future results. 

Investing involves risk, including the possible loss of principal and fluctuation of value. Nothing contained in this newsletter is an offer to sell or solicit any investment services or securities. Initial Public Offerings (IPOs) are highly speculative investments and may be subject to lower liquidity and greater volatility. Special risks associated with IPOs include limited operating history, unseasoned trading, high turnover and non-repeatable performance.

SITALWeek #416

Welcome to Stuff I Thought About Last Week, a personal collection of topics on tech, innovation, science, the digital economic transition, the finance industry, and whatever else made me think last week.

Click HERE to SIGN UP for SITALWeek’s Sunday Email.

In today’s post: OpenAI made it easy to create new custom GPTs, which allowed me to launch two beta chat agents; making better decisions by understanding everything that leads up to the crucial moment; signs that manufacturing capital is reshuffling dramatically around the globe; and, a bevy of birth rate stats as the hypothesized "sneaker wave" morphs into a receding rip tide. 

SITALWeek will return on November 26th.

Stuff about Innovation and Technology
Tailored Intelligence
Last week at OpenAI’s first developer conference, they announced the ability to create custom GPTs. The basic idea is that you task ChatGPT with a certain set of parameters and upload some supporting documents (or tie it to external APIs) to generate a customized chat agent. You can also share usage of the GPT (currently limited to GPT Plus subscribers). Soon, the company will be launching a GPT “app” store that will share revenues with creators of commercial GPT agents (i.e., if you create a GPT agent with access to useful proprietary data, it could become a revenue generating tool that can be used standalone or embedded in an application). Further, if you have an enterprise account with OpenAI, you can create proprietary GPTs (e.g., with access to internal data relevant to your company). This type of knowledge-base-on-steroids should help companies unlock productivity at an unprecedented pace. Another use case would be to build a customer service chatbot with access to your internal records. To experiment with these new features, I’ve launched a GPT called Rootmo that has embedded the knowledge of five of our NZS Capital whitepapers (GPT Plus subscribers can use it here; start with the recommended queries). I am also hard at work on BullSitter, the SITALWeek GPT (this GPT is incomplete, but you can try the sample questions). So far, I’ve found the GPTs to be rather limited because they are not indexing information to create a deep understanding; rather, they are just querying their uploaded documents. Thus, for now, you need to ask somewhat leading questions. For example, you can ask Rootmo to define ROOTMO for you; or, you can ask it how complex adaptive systems are relevant to investing. Despite OpenAI’s slow release of functionality, I think this customization advance marks a major step toward making AI more useful. Ethan Mollick has a more detailed writeup on these GPT “almost” agents. In other OpenAI news, the company announced that they have 2M active developers, 100M weekly active users, and over 92% of the Fortune 500 as customers; they have also slashed the developer pricing tier concurrent with the release of the new GPT4 Turbo. This combination of leading functionality and early traction is likely creating a network effect of such proportions that, with each passing day, it will be increasingly difficult for the competition to catch up; but, it’s early days and we should see more competitive models from Google and others soon.

Miscellaneous Stuff
Determined to Make Better Decisions
Robert Sapolsky has a new book out called Determined: A Science of Life without Freewill. I covered Sapolsky’s last book Behave in #210, quoting a passage that explains just how many things feed into our moment-to-moment conscious behavior: “blood glucose levels; the socioeconomic status of your family of birth; a concussive head injury; sleep quality and quantity; prenatal environment; stress and gluticocorticoid levels; whether you’re in pain; if you have Parkinson’s disease and which medication you’ve been prescribed; perinatal hypoxia; your Dopamine D4 receptor gene variant; if you have had a stroke in your frontal cortex; if you suffered childhood abuse; how much cognitive load you’ve borne in the last few minutes; your MAO-A gene variant; if you’re infected with a particular parasite; if you have the gene for Huntington’s disease; lead levels in your tap water when you were a kid; if you live in an individualist or collectivist culture; if your a heterosexual male and there’s an attractive woman around; if you’ve been smelling the sweat of someone who is frightened. On and on. Of all the stances of mitigated free will, the one that assigns aptitude to biology and effort to free will, or impulse to biology and resisting to free will, is the most pernicious and destructive.” I think being cognizant of the myriad factors and inputs that drive human behavior could have a huge positive impact on our decision making process (a topic I also covered in Time Travel to Make Better Decisions – a paper you can ask Rootmo about!). As much as I love Behave, it’s a bit long and dense and not everyone has the will to take the time to read it. In Determined, Sapolsky offers a condensed (but fairly comprehensive) summary of Behave before delving into the philosophical implications of how to live once you have internalized the ramifications of his research. I’ve discussed this illusory freewill in the past in posts such as this one. Whether or not you have read Behave, I recommend checking out Determined to better understand how the human mind functions.

Stuff About Demographics, the Economy, and Investing
Reglobalization
While attempts at reshoring/reorganizing global supply chains have proven slow and fraught with challenges, it appears a material shift is finally taking place. Foreign direct investment (FDI) in China went negative (outflows of capital) for the first time since China began tracking the stat. After averaging $50-100B per quarter in the 2010s, it recently dropped to negative $11.8B in Q323. Meanwhile, in Mexico, FDI hit an all-time record high of $33B for the first three quarters of 2023, which is ~30-50% above earlier 2020s levels. Around half the total FDI was directed towards manufacturing of all types, including a quarter slated for the automotive sector. On a related note, this WSJ article on America’s inability to manufacture something as simple as synthetic rubber gloves – despite a major effort and large amount of money – is depressing.

The Childless Generation
It’s rare that I come across a novel presentation of demographic stats, but the WaPo’s story on low US birth rates contains some eye-catching data. While Millennials are getting married and moving into houses, they still are not having babies. While it’s possible the numbers reflect delayed childbearing, even rising births for older women are not enough to offset the declining births for younger women. A recent update to the Census forecast (last done in 2017) now shows that there will be more deaths than births in the US just fifteen years from now. Previously, this was not forecast to happen until much later in the century. While there is ample evidence pointing to a long series of economic hardships that conspired against younger generations in the 2000s, even now, more and more young adults are zero-child households. As of 2022, almost 35% of women aged 25 to 44 have had no children, up from ~28% in 2012. The WaPo also cites a 2021 Pew Research poll that 44% of childless adults aged 18 to 49 were “not too likely” or “not at all likely” to have a child, up from 37% in 2018. Further, it’s harder for kids to out-earn their parents these days. For example, 90% of kids born in the 1940s were more prosperous than their parents, but that figure had dropped to ~50% by the time the first wave of Millennials were born in the early 1980s. According to the WSJ, over 40% of families headed by college grads born before 1968 have a net worth in excess of $1M. The remarkably lucky set of circumstances those generations experienced have yet to repeat. With current prospects for building wealth hampered by sky-high home prices, young adults are more likely to be pessimistic regarding their (prospective) progeny’s prospects. Whereas I once described (back in 2019) an imminent demographic sneaker wave of Millennial births, it appears the trend has applied more to household formation and less to births. While I do worry about a shrinking set of opportunities as AI envelopes many white (and eventually blue) collar jobs, I am equally as optimistic (for now) that technology will create an array of unexpected new opportunities that multiply the size of the overall economy despite the impending population decline. Afterall, if you’ve read Sapolsky, you know that I have no choice but to be optimistic.

✌️-Brad

Disclaimers:

The content of this newsletter is my personal opinion as of the date published and is subject to change without notice and may not reflect the opinion of NZS Capital, LLC.  This newsletter is an informal gathering of topics I’ve recently read and thought about. I will sometimes state things in the newsletter that contradict my own views in order to provoke debate. Often I try to make jokes, and they aren’t very funny – sorry. 

I may include links to third-party websites as a convenience, and the inclusion of such links does not imply any endorsement, approval, investigation, verification or monitoring by NZS Capital, LLC. If you choose to visit the linked sites, you do so at your own risk, and you will be subject to such sites' terms of use and privacy policies, over which NZS Capital, LLC has no control. In no event will NZS Capital, LLC be responsible for any information or content within the linked sites or your use of the linked sites.

Nothing in this newsletter should be construed as investment advice. The information contained herein is only as current as of the date indicated and may be superseded by subsequent market events or for other reasons. There is no guarantee that the information supplied is accurate, complete, or timely. Past performance is not a guarantee of future results. 

Investing involves risk, including the possible loss of principal and fluctuation of value. Nothing contained in this newsletter is an offer to sell or solicit any investment services or securities. Initial Public Offerings (IPOs) are highly speculative investments and may be subject to lower liquidity and greater volatility. Special risks associated with IPOs include limited operating history, unseasoned trading, high turnover and non-repeatable performance.