A Few Things: What Creates Investment Opportunities, Why Do We Believe Conspiracies, Commodities In An Age of Scarcity, What's Happening In AI, Science of Aging, Matt Levine on Crypto, VC Metrics....

October 31, 2022

I am sharing this weekly email with you because I count you in the group of people I learn from and enjoy being around.

You can check out last week’s edition here: Morgan Housel's Rules, Quality Demands Quantity, Clocktower on China, The Chip War On China, Coatue on Fintech, Balaji with Lex Fridman, Meltem on Creation and Monetisation of Culture

---

Quotes I’ve been thinking about:

“There are a thousand thousand reasons to live this life, every one of them sufficient”

- Marilynne Robinson 

“It’s so much easier to suggest solutions when you don’t know too much about the problem.”

- Malcolm Forbes

"We act as though comfort and luxury were the chief requirements of life, when all that we need to make us happy is something to be enthusiastic about."

- Priest and novelist Charles Kingsley on what makes us happy

"Busy is a decision. We do the things we want to do, period. If we say we are too busy, it is just shorthand for the thing being "not important enough" or "not a priority." Busy is not a badge. You don’t find the time to make things, you make the time to do things."

- Writer, artist, and podcast host Debbie Millman on busyness

---

A. A Few Things Worth Checking Out:

1. Neckar had a piece titled: What Creates Opportunity?, it’s a discussion about the game of investing and where opportunities come from.

He discusses a number of great investors and focuses on Michael Mauboussin’s BAIT framework:

“Who is on the other side,” Mauboussin’s paper on investment edge, should be required reading for any investor. “The key to winning,” he writes, “is participating in a game ... [as] the most skilled player.” However, markets are “generally highly competitive” and low-skill games only remain so if there are barriers for sophisticated players such as size constraints (the best investors quickly exit the micro-cap space by virtue of their success).

“If you buy or sell a security and expect an excess return, you should have a good answer to the question “Who is on the other side?” In effect, you are specifying the source of your advantage, or edge.

We categorize inefficiencies in four areas: behavioral, analytical, informational, and technical (BAIT).”

Informational: “Information inefficiency arises when some market participants have different information than others and can trade profitably on that asymmetry.”

Analytical: “Analytical inefficiency arises when all participants have the same, or very similar, information and one investor can analyze it better than the others can.”

Technical: “Technical inefficiency arises when some market participants have to buy or sell securities for reasons that are unrelated to fundamental value.”

Behavioral: People make mistakes, individually and as part of groups.

“Behavioral inefficiencies may be at once the most persistent source of opportunity and the most difficult to capture. Many behavioral inefficiencies emanate from the psychology of belief formation and the psychology of decision making. It is essential to remember that these inefficiencies are generally the result of collective, not individual, actions.”

2. Hidden Forces spoke to Michael Shermer. Michael is the Founding Publisher of Skeptic magazine. He hosts his own podcast, “The Michael Shermer Show,” and is the best-selling author of books like “The Believing Brain, “Giving the Devil His Due,” “The Moral Arch,” “The Mind of the Market,” and many more.

Shermer's latest book titled "Conspiracy: Why the Rational Believe the Irrational" presents an overarching review of conspiracy theories—who believes them, why, which ones are real, and what we should do about them.

Conspiracy theories have long been a fringe part of the American political landscape.

The purpose of this conversation is to provide you with a framework for thinking about conspiracies—what they are, the different types of conspiracies that exist, and why we believe in them.

You will also learn how to distinguish between real conspiracies and imagined ones and what we can do as a society to tilt the information landscape toward producing more accurate models of the world without resorting to censorship or the policing of thought and information.

I read Michael’s book on religion a few years ago and found it really deep:

3. Charlie McGarraugh and Michael Kao spoke to Mike Green about Commodities in an Age of Scarcity. Anytime Charlie speaks you have to listen. He was a top 5 smartest person I worked with at GS.

4. Jim Chanos had a fireside chat with Mike Green on markets:

---

B. What’s Happening in AI:

This weekend I got fascinated with Generative AI, and went down the rabbit hole of understanding both how it works and what else is happening in AI today.

If you just want to get an overview of the current state of the art in AI, listen to this conversation between Lex Fridman and Andrej Karpathy.

Andrej Karpathy is a legendary AI researcher, engineer, and educator. He's the former director of AI at Tesla, a founding member of OpenAI, and an educator at Stanford.

Key bits:

0:58 - Neural networks

11:32 - Aliens

33:34 - Transformers (the AI kind)

41:50 - Language models

1:05:44 - Software 2.0

2:36:23 - Advice for beginners

2:45:40 - Artificial general intelligence

3:04:53 - Future of human civilization

3:09:13 - Book recommendations

3:17:12 - Future of machine learning

To get your arms around what’s happening in AI, there are a few terms worth knowing:

Transformers: a neural network deep learning model / architecture to transform an input to an output usually composed of an encoder and decoder. They are a form orm of semi-supervised learning, that use an “attention” mechanism to understand the context of text inputs.

Quick 8 min video here:

Some teenagers are using Transformers like GPT3 to write their college papers.

The other key terms to know in AI is GAN’s.

Generative Adversarial Networks (GANs): a model with a generator and discriminator, where the discriminator's job is to be able to tell whether the thing it’s looking at is real / good or fake / bad. The generator’s job is come up with data / move / action to beat or fool the discriminator.

If the discriminator gets fooled, it updates itself to not be beaten next time. If the generator gets found out, it updates itself to do better next time. The model constantly battles it out and learns on its own.

This is how Deep Mind learnt how to play Chess and GO.

So far much of the innovation in AI has been inside the big tech firms: Google, Amazon, Tencent, Microsoft, even OpenAI is relatively closed…..This is probably not a good thing in the long run.

One of the companies at the forefront of AI which is trying to bring AI to the masses is Stability AI (just raised $100mm at $1bn from Lightspeed and Coatue).

This chat with the CEO and founder of Stability AI, Emad Mostaque discusses the emerging trends for open source AI infrastructure, generative models, the importance of data for real-world applications of AI, and predictions on the development of “text-to-everything” in artificial intelligence.

Back to Generative AI, which took me down this rabbit hole in the first place.

Generative AI is responsible for pictures like the one below, that my 11 year old made in 10 secs by asking DALL-E2 to draw a Polar Bear in Monet’s style.

Or check out this AI generated podcast of Steve Jobs on the Joe Rogan show.

Crazy!

Here’s some context on Generative AI: The BIG Think had an article titled: How do DALL-E, Midjourney, Stable Diffusion, and other forms of generative AI work?

Key bit on how diffusion models, which is a relatively new idea in AI (emphasis mine)

Generative Pre-trained Transformer 3 (GPT-3) is the bleeding edge of AI technology. The proprietary computer code was developed by the misnomered OpenAI, a Bay Area tech operation which began as a non-profit before turning for-profit and licensing GPT-3 to Microsoft. GPT-3 was built to produce words, but OpenAI tweaked a version to produce DALL-E and its sequel, DALL-E 2, using a technique called diffusion modeling.

Diffusion models perform two sequential processes. They ruin images, then they try to rebuild them. Programmers give the model real images with meanings ascribed by humans: dog, oil painting, banana, sky, 1960s sofa, etc. The model diffuses — that is, moves — them through a long chain of sequential steps. In the ruining sequence, each step slightly alters the image handed to it by the previous step, adding random noise in the form of scattershot meaningless pixels, then handing it off to the next step. Repeated, over and over, this causes the original image to gradually fade into static and its meaning to disappear.

When this process is finished, the model runs it in reverse. Starting with the nearly meaningless noise, it pushes the image back through the series of sequential steps, this time attempting to reduce noise and bring back meaning. At each step, the model’s performance is judged by the probability that the less noisy image created at that step has the same meaning as the original, real image.

While fuzzing up the image is a mechanical process, returning it to clarity is a search for something like meaning. The model is gradually “trained” by adjusting hundreds of billions of parameters — think of little dimmer switch knobs that adjust a light circuit from fully off to fully on — within neural networks in the code to “turn up” steps that improve the probability of meaningfulness of the image, and to “turn down” steps that do not. Performing this process over and over on many images, tweaking the model parameters each time, eventually tunes the model to take a meaningless image and evolve it through a series of steps into an image that looks like the original input image.

---

C. The Tech and Crypto Section:

1. MIT Technology Review had a discussion around Research labs pursuing technology to “reprogram” aging bodies back to youth.

This along the lines of David Sinclair’s work on epigenetics, which we covered here on the Science of Aging.

2. Bloomberg’s Matt Levine has delivered what is probably the most balanced, easy-to-understand yet comprehensive explainer of Crypto. It’s aptly titled “The Crypto Story You Need”, he takes the reader through the basic principles, technologies, controversies and innovations of the crypto world.

Whether you love crypto or think it’s a scam this is a piece worth reading. As long time riders will remember I spent a lot of time last year going down the rabbit hole, and this piece summarises what I spent months learning.

Here’s the beginning and the end to give you a flavour.

Introduction:

I need to give you some warnings. First, I don’t write about crypto as a deeply embedded crypto expert. I’m not a true believer. I didn’t own any crypto until I started working on this article; now I own roughly $100 worth. I write about crypto as a person who enjoys human ingenuity and human folly and who finds a lot of both in crypto.

Conversely, I didn’t sit down and write 40,000 words to tell you that crypto is dumb and worthless and will now vanish without a trace. That would be an odd use of time. My goal here is not to convince you that crypto is building the future and that if you don’t get on board you’ll stay poor. My goal is to convince you that crypto is interesting, that it has found some new things to say about some old problems, and that even when those things are wrong, they’re wrong in illuminating ways.

Also, I’m a finance person. It seems to me that, 14 years on, crypto has a pretty well-developed financial system, and I’m going to talk about it a fair bit, because it’s pretty well-developed and because I like finance.

And the end:

Crypto, meanwhile, has built a financial system from first principles, pure and pleasing on its own, unsullied by contact with the real world. (I exaggerate: The basic function of sending money using crypto, Satoshi’s original goal, is fairly practical. But, otherwise.) That’s interesting as an object of aesthetic contemplation, and I’ve enjoyed contemplating it, and I hope you have, too. And it’s attracted a lot of finance people who also enjoy contemplating it, and getting rich. And their task is to build back down, step by step, to connect the elegant financial system of crypto to the real world. You’ve built a derivatives exchange, cool, cool. But can a real company use it to hedge a real risk facing its real factory? You’ve built a decentralized lending platform, awesome. But can a young family use it to buy a house?

And the answer is, you know, maybe, give it time. The crypto system has attracted a lot of smart people who want to solve these problems, in part because they’re intellectually interesting problems and in part because solving them will make these people rich.

But another part of the answer might be that the real world—growing food, building houses—is a smaller part of economic life than it used to be, and that manipulating symbolic objects in online databases is a bigger part. Modern life is lived in databases. And crypto is about a new way of keeping databases (on the blockchain).

If you build a financial system that has trouble with houses but is particularly suited to financing video games—one that lets you keep your character on the blockchain, and borrow money from a decentralized platform to buy a cool hat for her, or whatever, I don’t know—then that system might be increasingly valuable as video games become an increasingly important part of life. If you build a financial system whose main appeal is its database, it will be well-suited to a world lived in databases. If the world is increasingly software and advertising and online social networking and, good Lord, the metaverse, then the crypto financial system doesn’t have to build all the way back down to the real world to be valuable. The world can come to crypto.

3. The All-In Podcast had an interesting episode discussing two key ideas:

(32:53) Macro outlook, Stability AI's $101M fundraise, VC fund metrics

The key chart that caught many people’s attention was - while VC mark-ups, TVPI is strong, how much of it will convert into DPI.

How and where will the two bars converge?

TVPI = total value to paid-in i.e. (current value of remaining investments within a fund + total value of all distributions to date) / total amount of capital paid into the fund aka “Unrealised + realised P&L”

DPI = Distribution to paid-in i.e. total amount of distributions to date / total amount of capital paid into the fund aka “realised P&L”

The 2nd part of the discussion was at (1:05:38) focused on the difficulty of active stock picking and valuing businesses.

On Venture, this is a great blog post titled: Venture Capital Is Ripe for Disruption.

for large VC funds, a startup achieving a billion-dollar outcome is meaningless. To hit a 3-5x return for a fund, a venture partnership is looking to partner with startups that can go public at north of $50B dollars... there are only 48 public tech companies that are valued at over $50B... there are close to 1,000 venture funds all trying to find these select few. This is a huge problem.

---

Desidirata by Max Ehrmann:

Go placidly amid the noise and the haste, and remember what peace there may be in silence. As far as possible, without surrender, be on good terms with all persons.

Speak your truth quietly and clearly; and listen to others, even to the dull and the ignorant; they too have their story.

Avoid loud and aggressive persons; they are vexatious to the spirit. If you compare yourself with others, you may become vain or bitter, for always there will be greater and lesser persons than yourself.

Enjoy your achievements as well as your plans. Keep interested in your own career, however humble; it is a real possession in the changing fortunes of time.

Exercise caution in your business affairs, for the world is full of trickery. But let this not blind you to what virtue there is; many persons strive for high ideals, and everywhere life is full of heroism.

Be yourself. Especially do not feign affection. Neither be cynical about love; for in the face of all aridity and disenchantment, it is as perennial as the grass.

Take kindly the counsel of the years, gracefully surrendering the things of youth.

Nurture strength of spirit to shield you in sudden misfortune. But do not distress yourself with dark imaginings. Many fears are born of fatigue and loneliness.

Beyond a wholesome discipline, be gentle with yourself. You are a child of the universe no less than the trees and the stars; you have a right to be here.

And whether or not it is clear to you, no doubt the universe is unfolding as it should. Therefore be at peace with God, whatever you conceive Him to be. And whatever your labors and aspirations, in the noisy confusion of life, keep peace in your soul. With all its sham, drudgery and broken dreams, it is still a beautiful world. Be cheerful. Strive to be happy