SITALWeek #355
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: complex adaptive systems teach us that winning non-zero-sum games and adapting to whatever comes is a path to progress, but many folks takeaway the wrong conclusions from the science; artists of all kinds are increasingly using AI-powered tools to aid in creation, including fiction writing; teaching AI to behave around humans can dramatically boost its effectiveness; a look at the high speed-data link on the Webb telescope; Microsoft laments the declining labor force, will they and other companies lobby for immigration?; extreme heat requires a lot of labor to fix and prepare for future events; and, much more below...
Stuff about Innovation and Technology
AI Co-Authors
Sudowrite is a tool built on the language model GPT-3 for writing fiction: “Authors paste what they’ve written into a soothing sunset-colored interface, select some words, and have the AI rewrite them in an ominous tone, or with more inner conflict, or propose a plot twist, or generate descriptions in every sense plus metaphor.” The Verge interviewed one author, Jennifer Lepp, who tried the tool out with interesting but mixed success, especially as Sudowrite kept inserting out-of-context references to swords. Sudowrite also has a penchant for overly extended metaphors and “hallucinatory weirdness”, somehow managing to connect elephants with sharks and a plane full of ninjas. But, as the author learned to prompt Sudowrite better (it’s “like trying to teach tricks to a superintelligent cat” according to another writer, Gwern Branwen), she used the tech to co-write more and more of her books, in some cases without significantly editing the outcome. Ultimately, after feeling alienated from the stories, Lepp took most of the writing back herself, but still often uses Sudowrite for smaller bits of text. Obviously, we are still figuring out how art and AI will intersect. The article notes that we’ve become accustomed to images being heavily manipulated by people and algorithms, so AI having a hand in generating text should, likewise, not be terribly hard to digest. As I noted in a previous post on Dall-E, it’s possible these tools will paradoxically give us more, not less, agency over our creative visions. For all you know, this newsletter is the result of me just plugging in some prompts and article links and hitting send from the beach. How would you know? In a world of copies of copies, are we still capable of identifying a truly original creation? Artists, like all tradespeople and engineers, have sought to incorporate useful new technologies to advance their work (checkout the documentary Tim’s Vermeer for an interesting example). Should we treat these new tools any differently than paint brushes or word processors? Regardless, given the rise of short-form content, who has time for detailed, immersive storytelling anyway? Clearly what we need is AI to generate TikToks.
AI Etiquette
Sony retrained its Gran Turismo video game racing AI engine on a larger neural network after it continually lost to human players, but the key to GT Sophy winning wasn’t just speed, it was etiquette. Once GT Sophy was better able to use context to know whether to be more aggressive or back off, it became hard to beat. Sony recently used the AI engine in the real world to best champion drone racers. The researchers believe that knowing how to “behave” will become a key attribute for any AI system that interfaces with humans.
Miscellaneous Stuff
Alzheimer’s Gut-Brain Connection
Researchers at Edith Cowan University in Australia have found a widespread genetic link between gut disorders and Alzheimer’s Disease, with cholesterol levels potentially acting as a contributing factor. “Looking at the genetic and biological characteristics common to AD and these gut disorders suggests a strong role for lipids metabolism, the immune system, and cholesterol-lowering medications. Whilst further study is needed into the shared mechanisms between the conditions, there is evidence high cholesterol can transfer into the central nervous system, resulting in abnormal cholesterol metabolism in the brain. There is also evidence suggesting abnormal blood lipids may be caused or made worse by gut bacteria (H. pylori), all of which support the potential roles of abnormal lipids in AD and gut disorders. For example, elevated cholesterol in the brain has been linked to brain degeneration and subsequent cognitive impairment.”
Webb’s Data Link
Last week, I linked to some resources for viewing the detailed images coming back from the Webb telescope. IEEE had some interesting stats about the generation and transmission of raw data (that researchers turn into the stylized and colorized final products) from the JWST. Data arrive via a 28mb/s link in the Ka-band at 25.9 GHz. The Webb can produce up to 57GB of data per day depending on its schedule. The uplink is quite a bit slower – 16kb/s – for sending instructions to the scope in the 2.09 GHz S-band. Webb also has a 68GB solid-state drive for data storage when it’s out of direct contact with the Earth. At the speed of light, transmissions take a mere ~5 seconds to traverse the ~1 million miles of space separating the Webb from Earth. So, the light from galaxies in the distant past traveled for 13B years to land on the Webb, and a mere 5 more seconds to arrive on Earth in the form of 1s and 0s. NASA makes the data from the telescope available for anyone to process, and last week amateur space image processor Judy Schmidt posted this amazing image of the spiral galaxy M74.
Stuff about Geopolitics, Economics, and the Finance Industry
Growing Awareness of Labor Stagnation
Microsoft’s president Brad Smith lamented the growing dearth of kids entering the labor force in developed countries in a Reuters interview. This ebb is of course no surprise to SITALWeek readers who are surely tired of me (or the AI that writes this newsletter) talking about the consequences of the declining birth rates combined with lower immigration. For Microsoft, that means fewer future Office 365 subscribers. For businesses in general, the trend entails a future of lower growth offset by increased automation and software-driven productivity. You don’t have to look too far to see a probable wave of deflation resulting from an aging population and accompanying technology investments. Declining population growth is akin to losing a principal economic engine and effectively transitioning from a non-zero to a zero-sum economy. For example, without enough new consumers being born into the world, in order for Coke to grow volume, all things equal, it must take share from Pepsi. It’s been hundreds of years since we’ve dealt with a zero-sum economy (or, worse, a negative-sum economy where all output went to the king or state). That said, there are other ways to grow an economy with a flat consumer base, such as productivity gains, innovation, and government deficit spending. (The latter mechanism is the least desirable and most tenuous, but it has perhaps been the primary engine of growth for the last few years, and certainly was during the pandemic). To the extent that more large companies like Microsoft realize how shrinking labor markets will lead to a zero-sum economy, I expect we could see a large lobbying effort to increase immigration. As I’ve said before, the countries with the most attractive immigration policies to offset their demographic headwinds will enjoy the strongest economies in the coming decades. Of course, immigration itself is zero sum, with one country’s gain being another’s loss.
Climate Damage Repairs Face Labor Headwind
Extreme heat is taxing a world built for lower temperatures. Melting roads and runways, warped train tracks, and data-center cooling failures are just a few examples. Some of these issues – like improving data-center cooling – can be solved with technology and AI, but most of the problems require expensive, labor-intensive efforts. The demographically shrinking labor force in developed countries means that we might need to simply adjust to a world with Internet outages, unreliable transportation infrastructure, power rationing, etc. From #310:
As I was thinking about the cost to upgrade infrastructure to handle more extreme weather swings, it seems like there are a lot of ~$20B projects under consideration. A couple weeks ago, I mentioned (#306) it would cost about that much for PG&E to bury a portion of its power lines in high risk areas, and for Detroit to upgrade their stormwater drainage system. Apparently, it was also determined that damming the Golden Gate Bridge to keep rising tides at bay would cost...$19B. It’s certainly easy to see how expenses could add up to well into the trillions. I can’t help but wonder where the labor will come from (and with what incentives) to even consider some of these projects. 21,000 people were involved in the construction of the Hoover Dam. Is it less people-intensive to build a dam today than during the New Deal era? It’s possible that governments won’t even be able to contemplate breaking ground until deflationary automation/robotics renders construction more affordable.
In the long term, we will solve these problems through technological innovation, but, in the meantime, try to stay cool and pray that TikTok has good cooling for its servers.
Optimism Inherent to Complex Systems
I highly recommend Niall Ferguson’s Bloomberg op-ed on complex adaptive systems. It’s a great overview even if you already understand the science. While he nails the key ideas in detail, he ends on a misinformed, cynical note: “There can be, and often are, cascades or chain reactions of disaster. The more networked the world becomes, the more we see this.” I reject that notion, but I’m also not trying to sell a book titled Doom: The Politics of Catastrophe. There are surely some examples where his thesis is correct, but if the assertion were true generally, then evolution’s increasingly complex network of organisms would have failed long ago – we simply wouldn’t be here (or maybe we are just spectacularly lucky). If anything, the opposite is true: the hallmarks of survival and success in complex systems are adaptability, progress via non-zero-sum game outcomes, and innovation – all things at which humans excel. Increasing the number of nodes in the network, or the speed with which they communicate, does not automatically lead to one disaster begetting another. We might experience more extreme, “fat-tail” events within the chaotic conditions of an interacting network of agents, but even those shocks to the system have not yet managed to collapse the planet or derail our development as a species. Instead, progress in complex adaptive systems tends to come from redundancies, problem solving, and faster adaptations.
In contrast to Ferguson’s doom and gloom, the functioning of our world as a nested, multifaceted complex adaptive system is significant reason for optimism. Expecting the worst is a classic, frequent mistake – it’s natural that the more information you have and the more clever you are, the more negative examples and patterns you might see, and our brain tends to accord more weight to the negatives to increase the odds of our survival. The seductiveness of pessimism and cynicism is hard for many to avoid. However, since The Scientific Revolution, which began around 500 years ago, these fearful takes on the world have proven to be broken long-term philosophies. My outlook is not a rose-colored Pollyanna vision. To be sure, we have some epic challenges in the world today, and a lot of things could get worse before they get better. However, complex adaptive systems give us the tools to plot a course to credible success: build resilience, adapt, and create non-zero-sum (win-win) outcomes. We can most effectively solve problems by preparing and innovating for an array of outcomes rather than specific, narrow visions of the future. The paradox of Ferguson predicting cascading failures is that the key lesson from complex adaptive systems is we cannot predict anything, good or bad.
Complexity Investing is our synthesis of how we relate the ideas from complex adaptive systems to investing. We wrote the first draft of the paper in the fall of 2012 (nearly ten years ago!), and we are always excited when complex systems gain notoriety because we think they are the model for understanding the world. However, mentally shifting to a complex systems framework requires that you give up the idea that you can predict the future with any meaningful degree of precision, which is a considerable challenge. Consequently, the vast majority of investors who have read the paper have not fully grasped its implications.
In the spirit of increasing awareness of complex adaptive systems, we did something slightly unusual this week by having our whitepaper, Complexity Investing, sponsor The Acquired podcast. The episode is a deep dive on Walmart, with some fascinating history (that I certainly wasn’t fully aware of). Our goal wasn’t (and probably never will be!) to advertise NZS Capital. Rather, we hope our sponsorship brings in some new viewpoints and spurs some lively discussion and debate on the ideas we’ve set forth in Complexity Investing and our other papers (all of which can be found here). We’ll also be doing a live Zoom with Ben and David on August 16, 2022, for which you can register via the episode link above if you would like to ask questions or tell us where we may have gone wrong. Or, shoot us an email anytime – AI-generated text is encouraged.
✌️-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.