trevor (taylor’s version)
@tmychow
diffusing the agi
openai launched gpt-4.5 to great excitement yet under 2 months after launch, they announced it would be deprecated is pre-training over now? @jackgwhitaker and i show: 1. pre-training scaling laws haven't bent 2. why the marginal dollar left it for RL 3. why it'll come back
so excited this post is finally out! capital markets are a phenomenal instruments for shaping incentives and transferring risk sensibly, and that includes risks from the deployment of ai models
@RajivDattani, @bradr and I spent the last year exploring how insurance can unlock secure AI progress Informed by our time at Anthropic, McKinsey insurance, METR, Center for AI Safety Our essay: underwriting-superintelligence.com/?july
there’s something electric about being on the trading floor. at prev job (opti/drw/jane) i vividly remember the first time walking onto the desk: flow scalpers rolling in, 9 monitors, positioned so your field of view was pure screen. a cocoon taking you to the world of convexity
Sheryl (@sherylhsu02) was our first hire onto the multi-agent team. Within a few months of joining, she helped to make this possible. We're so lucky to have her on the team!
Watching the model solve these IMO problems and achieve gold-level performance was magical. A few thoughts 🧵
last summer, alphaproof + alphageometry2 combined to achieve a silver at the IMO yet @polymarket was only pricing a 25% chance of AI models achieving a gold this year this year's IMO just occurred, and @openai smashed it with a single model without tool use and got a gold!
1/N I’m excited to share that our latest @OpenAI experimental reasoning LLM has achieved a longstanding grand challenge in AI: gold medal-level performance on the world’s most prestigious math competition—the International Math Olympiad (IMO).
"Madam Chair, I would like to enter into the record Apollo Research’s report titled *Frontier Models are Capable of In-Context Scheming*" is not a sentence I would have expected this early in the timeline
Absolutely phenomenal exchange from @RepScottPerry at a recent US House hearing about AI – demonstrates way more understanding of the issue than is typical: Rep Perry: “I do want to make everybody aware of some things that I am aware of. I’m going to refer directly to an AI…
these results were eye-opening for me... chatgpt agent performed better than i expected on some pretty realistic investment banking tasks
i really hope that one day we'll get a book which goes into Goldman risk-arb (Rubin era) and Goldman SSG + how they each, in their own era, produced an incredible track record of alumni
In it’s heyday, Goldman’s special situations group was the navy SEALs of money making. Today’s guest, Alan Waxman, used to run that group before leaving to build Sixth Street, now a $115b behemoth. SSG heads were sometimes of the more brainy, nerdy variety. Not Waxman. He is a…