Trying to stay sane

This post could much more accurately be titled as ‘The exquisite agony of being faced with possibly the greatest technology shift we’ve ever known just as a bear market starts to open up some value-investing opportunities’.

There are several obvious shifts, rotations and pivots (including some which may actually be death spirals in disguise) happening which I’ve been attempting to stay current with, a non-exhaustive list includes:

• Bear market technology valuations
• The rise (and economic inversion) of creator platforms
• Toy companies transforming into…something
• Mono-polar geopolitics (and currencies) shifting to multi-polar
• Venture capital model evolutions
• The rise of Generation Alpha

All of these phase-shifts are being interrupted by the explosion of AI models.

The rate of change happening in AI is unlike any other tech trend we’ve seen. In the 90s we had to wait for everyone to get on the web. In the 2000s we had to wait for everyone to get smartphones. Now any progress spreads instantly to all users and developers.

— Aaron Levie (@levie) May 16, 2023

This is a mania I instinctively distrust combined with a technology that I can’t dispute. It makes me feel pretty unequipped to evaluate most opportunities right now. 

I suspect between both of these questions there’s at least ten billion dollars in new value waiting to be created pic.twitter.com/z8Sp5U0EET

— Dylan Collins (@MrDylanCollins) May 7, 2023

I don’t think this is just me? Quite a few repeat founders I know who were mid-leap with their next business have pulled back and are just watching (and tinkering) to see how everything AI plays out. Several others are on their second or third pivot and seemingly in permanent reaction mode. Others are just ignoring it. 

I’ve been trying to update my mental models in the midst of this maelstrom. Some of my updated current priors (or it turns out more correctly, credences):

• The future doesn’t always come with good business models. Even assuming that AI can, at its extremes in certain industries, substitute humans entirely it still can’t replace the viability (or not) of the business model itself. 
• Startups with AI-enhanced offerings do not guarantee management excellence to run them well as companies 
• No level of AI will change the vast universe of human psychological biases which are interacting with the end-product (this is an interesting point when you think about the impact of AI on high switching-cost situations) 
• In a world with relatively unconstrained access to LLM models, there may be some interesting reversion-to-mean factor at work (at east on some timeline)

“There’s somethin’ about all this madness that I find calmin'” (Potter Payper, Blame Brexit)