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.