WSJ article came out yesterday: https://www.wsj.com/tech/ai/the-ai-revolution-is-already-losing-steam-a93478b1

Here’s to hoping

This feels a little too low to me. We certainly haven't entered such a winter yet, but we do seem to be starting to see some diminishing returns potentially. It's been 3 years and they still can't draw a pentagon.

It's possible but with so much investment from large private technology companies, nation states, public utilities; I just don't see it happening. Even if new features stall they will add compute and energy until the next breakthrough.

@LeeWoods I disagree, and this is pretty much the crux for me. The money being pumped into AI needs to see some sort of return. Increasing training costs x10 isn't worth it anymore, it's extremely hard to increase revenues proportionally.

See OpenAI and their soft pivot (if you want to call it that) to mundane utility. With omni, they're making a push towards commercialization at current AI tech levels. It should also be clear at this point that SORA was announced for the purpose of shopping it around to potential business partners in Hollywood. I suspect that generation is too expensive to offer it at reasonable rates to random consumers (this was already my guess based on my domain knowledge, but now we have practical evidence as well)