Current Micro-Season

Loading...

Loading...

Loading...

Loading...

Given all this, the fraction of meaning in the autocomplete view of current AI is alarmingly akin to the random, not always incorrect observations about temperature cycles conservatives used to throw around in debates about climate change. In both cases, a debatable description of mechanism is mistaken for proof of (in)significance. CO2 makes up only 0.04% of the atmosphere, which sounds much too little for it to drive global warming — until you learn CO2’s molecular structure lets it absorb infrared radiation in ways nitrogen and oxygen can’t. Similarly, “AI just predicts the next token” sounds deflating — until you consider what predicting the next token involves and start to ask if there’s really such a difference between predicting and learning.

I've been waiting for a bigger outlet to make this argument, that I've been making to anyone who will listen for a while now. Dan Kagan-Kans lays out an absolutely disheartening view of the anti-AI left and how they are willfully sitting out a transformative moment in society, when it matters most, because they've refused to actually engage with the topic beyond philosophical and definitional distinctions.

The parallels to climate change denial are unmistakable and make no sense this time either.

The continued use of the crypto bubble / hype cycle as a reason that tech can’t be “trusted” to be honest about AI is so bizarre to me. At the height of the crypto hype cycle, the vast majority of tech companies had nothing to do with crypto.

Aside from Meta, none of the big players got involved much at all besides perhaps some exploratory investments that were insignificant to their balance sheets (as big companies will do, just to see if there’s anything there).

Google, Microsoft, Amazon, etc. were not betting any part of the business on crypto. Based on my own experience and conversations, crypto boosters tended to be confined to rather small pockets at most companies.

I know tech can seem like a monolith to outsiders, but “crypto will change everything” was never a majority view. It just doesn’t make sense to draw these comparisons in good faith.

More concretely, not taking AI seriously might blind the left to its political uses. “One possible concern might be the left-wing abstaining from using the tools when the right-wing does not, in politics, campaigning, policy,” Bruenig worried. There is already some data to this effect: 44% of Republican political consultants use AI for work daily, compared to 28% of Democratic ones, according to the American Association of Political Consultants.

Indeed. I remember interviewing with a prominent Democratic-aligned PAC staffed with lots of former Obama for America veterans. I was told at the time by a very senior person in that organization that technology and data were fundamental edges for the Democrats, and that the Republicans would be unable to catch up. Obviously that wasn't true. I pointed out how mistaken this was at the time because of it was clearly the case that the Republicans would find technologically savvy operators to work for them and wouldn't ignore what was clearly a political advantage waiting to be applied. (I didn't get the job). How the tables have turned.
← All notes

Send your thoughts

Name and email are optional.

ESC
Type to search...
↑↓ to navigate to select