Joe Benton

@JoeJBenton

Soon joining @METR_Evals | Previously Alignment at Anthropic, PhD at University of Oxford

Berkeley, CA

joejbenton.com

Joined February 2017

63 Following

13.1K Followers

Replies Reposts Media

Pinned
Joe Benton
@JoeJBenton
Sep 11
I left Anthropic's safety team two weeks ago. Now feels like a good moment to explain why. AI companies are racing to build machines that are much smarter than any human, and we may not survive this. I want to work from the outside to ensure the public is informed about these
jbenton1.substack.com
Why I left Anthropic’s safety team to hold AI companies accountable
Two weeks ago, I left the safety team at Anthropic.
1.3K
Joe Benton
@JoeJBenton
22h
Seems like it would be good for the public to have more reliable ways to know whether superintelligence is imminent than AI companies vaugeposting in their IPO prospectus.
Daniel Kokotajlo
@DKokotajlo
Nov 3
We at AIFP are spooked by this information. IF (big IF) Anthropic really is imminently about to meet the RSP threshold of 2x acceleration over baseline, then it seems timelines to full RSI and superintelligence may be very short indeed. For example if we naively optimize the AIFM
Quote
Theo Bearman
@theobearman
Nov 2
+2
Anthropic updated its S1 this morning. Notable additions re automated AI R&D and human oversight of it: "[...] our internal estimates and our expectations regarding future models, rather than our current models, indicate that we may imminently meet our Responsible Scaling
8
Joe Benton
@JoeJBenton
Sep 9
This feels like a broadly accurate description of the situation: our industry may be on track to build systems that impose an unprecedented amount of risk on the world. Evaluating those risks, and changing incentive structures so we can address them, is why I'm joining METR.
Jacob Coxon
@hilbertspaess
Sep 9
I resigned from Anthropic today. I spent the last three years doing pretraining research at both OpenAI and Anthropic. Neither company is acting responsibly. They are racing straight to self-improving superintelligence and gambling with our lives. More thoughts below.
19
Joe Benton
@JoeJBenton
Sep 7
📄 As AI systems get increasingly powerful, it might be very important that they can help with key tasks like AI safety research. I think TASTE, this eval which @hasanbaig8 painstakingly built, may be one of the best signals of whether we're on track for our AIs to be helpful.
Hasan
@hasanbaig8
Aug 29
As we increasingly automate AI R&D, can we also automate AI Safety research? We built TASTE (The AI Safety Taste Evaluation), a benchmark to measure whether models can predict which research proposals experienced AI safety researchers prefer.
3
Joe Benton
@JoeJBenton
Sep 2
An exciting personal update: Last week I left Anthropic to join @METR_Evals to work on embedded assessment of AI risks. Anthropic has been great to me. I’ve always known, though, that if something more impactful came up, I’d move on.
54

Joe Benton

Joe Benton’s posts