Anthropic CEO Calls for A.I. Slowdown

9 pointsposted 5 hours ago
by cmiles8

4 Comments

cmiles8

5 hours ago

Honestly the PR cycles these last new weeks look like a thinly veiled attempt to take attention off the fact that none of these AI labs have a viable proposition to justify the valuation they desperately need to keep going.

Option A: Face the music that the financial model of these labs is completely unsustainable and trillion-plus dollar valuations are a pipe dream by any normal standards of business.

Option B: Say “hey because of safety we had to slow everything down” and use that as cover for the next 24 months as a convenient excuse for why the financial metrics are so terrible. I expect a lot of “we’d be killing it right now financially but destroying our financials was the right thing to do for humanity” type BS arguments in the coming months.

This looks very much like Option B.

steammaho

5 hours ago

Excuse for what? Their financial metrics are ok. Open ai could do this but not anthropic

PandaRider

5 hours ago

I didn't read the essay [1] in full but the answer to the question of "why now" is the OpenAI-HuggingFace incident. And the NYT article mentioned Jacob Coxon quitting...

The target audience of the essay seems to be pressuring OpenAI. I don't know if OpenAI will listen.

[1] https://darioamodei.com/post/we-must-pace-the-frontier