OpenAI: a method to the madness?

Maybe Altman and his minions actually know what they're doing

This is a special post: a two-part collaboration with AI Realist by the indomitable Maria Sukhareva. The subject? Why, of course the hundreds of billions changing hands around OpenAI.

The other side of our duet is here: Scaling the Hype: NVIDIA × OpenAI × Oracle
In her post, Maria takes the opposite side of the argument and analyzes problems around the three-side deal. Make sure to read her post as well for the full picture - and naturally subscribe to my esteemed interlocutor:

Avoiding the news about OpenAI has become increasingly difficult over the last few weeks years: ChatGPT, the soap opera of rise, and fall, and rise again of Sam Altman, the promises of AI utopia (with Universal Basic Income), the frenemy relationship with Microsoft… But let me tell you: over the last two weeks, they’ve really outdone themselves. They cut a deal with Nvidia ($100 billion worth of compute), Oracle (300 billion) - and there is talk building A LOT of data centers to power all this expansion. And by a lot, I mean: power consumption that’s comparable with entire countries.

Artistic representation by Sam Altman message to the people of Earth.

If you were expecting a skeptical take on the whole thing, you can stop reading now - for two reasons. First, because Maria wrote a brilliant one, and second: because the whole thing got me thinking.