🖙Today in Silicon Valley Brain
I had the misfortune of having a dumb take retweeted into my feed. Now you get to enjoy it!
The piece, "It is all about the superstars" is a prime example of inverted logic. Perhaps the best illustration is the part where they bemoan the lack of ability for EU corporations to "respond to AI" due to labor laws.
In the U.S. firms have been decisive:
“The rapid success of ChatGPT triggered immediate responses: Microsoft streamlined its workforce, invested $10 billion in OpenAI, and more in its own AI infrastructure. Meta paused its metaverse efforts, laid off 20,000 employees within months, and boosted its AI investments, spending a whopping $37 billion on computing infrastructure in 2024. 41 Similarly, Google, facing challenges in search, halted major projects, laid off 12,000 employees, and accelerated on AI by ramping up its R&D investments to $43bn in 2023, including hiring tens of thousands of engineers with AI background.”
You may have noticed that a lot of "responding to AI" is laying people off. Even if we ignore the human cost of flooding labor markets with tens of thousands of employees, the stress it causes each of them, and the fact that they'll have to compete against each other 1, does this make a lot of sense? Even if you think a lot of AI is useful, there's still a bubble. Also, where are you going to find all these "engineers with AI background"? Wouldn't it make more sense to train some of your existing staff on what is apparently an extremely lucrative new field rather than try to find a bunch of new engineers suddenly? Also, who is going to test, document, market, and sell the AI those engineers are creating (not to mention developing the non-AI parts)?
It's telling that Meta's "metaverse efforts" only get a passing mention. The metaverse was a much earlier bubble, and it has amounted to even less than the current round of AI. How much human labor was wasted by both Meta pursuing a dud and and rapidly pivoting to and then from said dud?
We also shoot right past Google "facing challenges in search." Many people have bemoaned the drop in quality of Google. It's pretty subjective and hard to measure, but it definitely seems to have gotten worse. While Google might think AI will address their issues in search, I'm more skeptical. And even if it does, it's not necessarily going to be a fast process. Also, surely having the ability to tap into a very powerful conventional search would be a powerful way to distinguish your AI from everyone else's?
Some people have attributed large layoffs to companies realizing they've been hiring too aggressively or wanting to cut costs. This would transform this article's premise "weaker labor protections are good so large companies can be more nimble" to "weaker labor protections are good so large companies can fix their hiring mistakes," which makes the argument seem even weaker.
Perhaps the author, Pieter Garicano, would counter that these are irrelevant to his point about superstars being key to the economy. But I would argue it's relevant to whether these superstars are actually worthy of the name. If the "superstars" are making a huge mistake, isn't it actually good to have something preventing them from suddenly laying everyone off?
The piece is somewhat ambivalent, seemingly acknowledging some of the obvious downsides of the American system, but it seems strikingly credulous of capitalist narratives to just assume large layoffs are good for innovation.
This is relatively small in the grand scheme of the U.S. economy, where in a typical month, millions of workers quit or are laid off, which is currently balanced out by a slightly larger number of people being hired. However, for a specific job (mostly software developers and a few related positions) in a specific place (mostly Silicon Valley), it's substantially tougher. Obviously, as workers go, many (most? all?) of those laid off are extremely privileged. Still, this sucks for these people!↩