Ari Seff (@ari_seff) header
Ari Seff (@ari_seff) profile picture

Research @OpenAI | Previously @Waymo, PhD @Princeton

Joined March 2017
Photo by Ari Seff (@ari_seff)Photo by Ari Seff (@ari_seff)Photo by Ari Seff (@ari_seff)Photo by Ari Seff (@ari_seff)Photo by Ari Seff (@ari_seff)Photo by Ari Seff (@ari_seff)Photo by Ari Seff (@ari_seff)Photo by Ari Seff (@ari_seff)Photo by Ari Seff (@ari_seff)Photo by Ari Seff (@ari_seff)
Can we model driving similarly to how we model language? Excited to introduce MotionLM, where we represent driving scenarios as conversations among road users. It establishes a new state-of-the-art in multi-agent motion forecasting: arxiv.org/abs/2309.16534
13
105
15
861
176,876
It would be semi-ironic if the "GPT moment for robotics" is... just GPT lol.
GPT-6 Astra can solve some seriously tricky spatial constraint tasks: from unlocking interlocked parts to threading a rope through three rings. Try it yourself now 👇 qinengwang-aiden.github.io/d…
19
42
3
2,264
185,298
Ari Seff retweeted
Announcing that GPT-6 Astra has pushed the prime gap to 186, with Lean formalization! I was 9 when I first heard the twin prime conjecture. Its elegance and Yitang Zhang’s legendary story have always stuck with me. A truly surreal night, being the first to see our model make progress, pushing 246 all the way down to 186, on a problem I’ve revered since I was a kid. For me, it felt like witnessing a new era of intelligence being born, made possible by everyone at @OpenAI!
56
242
43
2,073
153,071
I am not ready to declare we reached AGI unless I can buy a bot that empties my dishwasher. (And now I feel very proud every time I do it because looks like it is much harder than getting the Fields medal)
16
26
4
416
31,123
yes, nonsofic groups exist: this statement is one of many new beautiful results proved by Astra, our next major model. We're releasing 10 such Astra proofs, complete with lean certificates and CoT walkthroughs for each of them. The results are wide-ranging, from von Neumann algebras (disproof of Connes' Rigidity Conjecture) to better bounds for high dimensional sphere packing, for circuit complexity, for monochromatic triangles in multicolored graphs, and more. More thoughts here: openai.com/index/ten-advance…
277
940
538
6,638
4,291,572
Ari Seff retweeted
GPT-5.2 Thinking evals
328
465
472
3,741
2,031,898
I’m at ICCV this week in Hawaii. Excited to talk with folks about robotics, agents, and multimodal reasoning. Reach out to chat!
11
1,102
Ari Seff retweeted
watching chatgpt agent use a computer to do complex tasks has been a real "feel the agi" moment for me; something about seeing the computer think, plan, and execute hits different.
1,098
746
278
12,335
4,229,216
One of the most striking, non-text AI plots I've seen since ChatGPT launched. Scaling keeps working, this time for Waymo's tooling.
4
22
1
181
15,519
o3-pro: How many of each letter are there in the SHA1 of the answer to this question?
30
34
11
866
150,526
Ari Seff retweeted
Operator is now powered by o3, improving overall task success rate. Also results in clearer, more thorough, and better-structured responses.
Operator 🤝 OpenAI o3 Operator in ChatGPT has been updated with our latest reasoning model. operator.chatgpt.com/
45
63
12
997
163,482
we’ve trained our reasoning models to get better at using computers, and now operator is powered by o3. super fun effort with the team! computer use remains a tough frontier problem, but this update represents a substantial increase in capability: nitter.cf/OpenAI/status/19259630…
Operator 🤝 OpenAI o3 Operator in ChatGPT has been updated with our latest reasoning model. operator.chatgpt.com/
3
1
25
4,878
Ari Seff retweeted
fun watching people react to operator. reminds me of the chatgpt launch!
1,110
320
195
10,183
2,043,857
Ari Seff retweeted
Announcing The Stargate Project The Stargate Project is a new company which intends to invest $500 billion over the next four years building new AI infrastructure for OpenAI in the United States. We will begin deploying $100 billion immediately. This infrastructure will secure American leadership in AI, create hundreds of thousands of American jobs, and generate massive economic benefit for the entire world. This project will not only support the re-industrialization of the United States but also provide a strategic capability to protect the national security of America and its allies. The initial equity funders in Stargate are SoftBank, OpenAI, Oracle, and MGX. SoftBank and OpenAI are the lead partners for Stargate, with SoftBank having financial responsibility and OpenAI having operational responsibility. Masayoshi Son will be the chairman. Arm, Microsoft, NVIDIA, Oracle, and OpenAI are the key initial technology partners. The buildout is currently underway, starting in Texas, and we are evaluating potential sites across the country for more campuses as we finalize definitive agreements. As part of Stargate, Oracle, NVIDIA, and OpenAI will closely collaborate to build and operate this computing system. This builds on a deep collaboration between OpenAI and NVIDIA going back to 2016 and a newer partnership between OpenAI and Oracle. This also builds on the existing OpenAI partnership with Microsoft. OpenAI will continue to increase its consumption of Azure as OpenAI continues its work with Microsoft with this additional compute to train leading models and deliver great products and services. All of us look forward to continuing to build and develop AI—and in particular AGI—for the benefit of all of humanity. We believe that this new step is critical on the path, and will enable creative people to figure out how to use AI to elevate humanity.
5,283
10,100
8,068
58,146
33,960,931
Discrete generative models use denoisers for generation, but they can slip up. What if generation *isn’t only* about denoising?🤔 Introducing DDPD: Discrete Diffusion with Planned Denoising🤗🧵(1/11) w/ @junonam_ @AndrewC_ML @HannesStaerk @xuyilun2 Tommi Jaakkola @RGBLabMIT
6
56
5
233
40,690
Evaluating autonomous vehicles in simulation is only useful if the other agents exhibit realistic reactions. We introduce a simple scheme for running RL on top of pre-trained behavior models to improve agent realism. (new paper with my former colleagues at @Waymo):
1
10
1,580
Very cool postdoc opportunity at Princeton. Had the great pleasure of working with Ryan during my PhD. If you're an AI researcher considering a postdoc, highly recommend taking a look:
We are looking for AI Postdoctoral Fellows! Be a part of AI for Accelerating Invention, a new research initiative out of the Princeton AI Lab. You can learn about AI for Accelerating Invention here: invent.ai.princeton.edu/ and apply for the postdoc here: puwebp.princeton.edu/AcadHir….
2
1,105
Let's say we begin increasing the number of heads required to win (to 3, 4, 5, ...). When, if ever, does Alice first take the advantage? (poll below)
Flip 100 coins, labeled 1 through 100. Alice checks the coins in order (1, 2, 3, …) while Bob checks the odd-labeled coins, then the even-labeled ones (so 1, 3, 5, …, 99, 2, 4, 6, …). Who is more likely to see two heads *first*?
6%Alice
11%Bob
69%Equally likely
13%Don’t know/see results
14,682 votes • Final results
1
1
4
6,739
Alice is first favored at:
0%13 heads
78%26 heads
11%51 heads
11%Never
18 votes • Final results
3
1,040
After 3 incredible years at Waymo building generative models for driving, I've decided to take on a new challenge. I recently joined Apple to work on some ambitious AI initiatives. Excited for what comes next.
4
34
4,096
A huge thanks to my former Waymo colleagues for the wonderful collaborations. Waymo continues to be the clear leader in autonomous driving as the service expands to new cities. Can't wait to see what the team accomplishes going forward.
5
886