President & Co-Founder @OpenAI
Joined July 2010
- Tweets 4,815
- Following 65
- Followers 497,251
- Likes 1,386
Excited to compete in my first strength competition on March 16th in Santa Clara, in an expo with over 3k athletes and 10k audience members: facebook.com/events/21722260…! Been training hard, just like our reinforcement learning agents.
3
1
1
40
It’s never been more fun to be a reinforcement learning agent.
We’re releasing a Neural MMO — a massively multiagent game environment that supports numerous populations of agents: blog.openai.com/neural-mmo/
- Code: github.com/openai/neural-mmo
- 3D Client: github.com/jsuarez5341/neura…
2
25
3
191
Amazing to see OpenAI featured twice on @BillGates's 2019 list of Breakthrough Technologies (once for robotics, once for language): technologyreview.com/lists/t…. Congrats to the team!
2
19
136
One very important aspect of OpenAI culture!
The aspect of @OpenAI I've been most impressed with since joining is the aura of cooperation. I've seen very few cases of people prioritizing themselves before others, but many cases of people helping with each other's project without obvious personal gain
7
65
Enjoyed watching last night's conversation with @AmandaAskell @AnimaAnandkumar @Miles_Brundage @smerity @WWRob. Very glad to see productive debate about publishing for GPT-2 and future powerful models (in a more nuanced way than is possible on Twitter):
Tonight we held the 1st ever TWiML Live, an engaging discussion on the controversy behind the release of @OpenAI's new language model. To learn more about this model, the concerns raised by its release, and how we can address them, don't miss the replay at twimlai.com/live-openai
2
13
51
If you’re a social scientist, please consider applying to OpenAI!
11
3
77
Agree that the implications are scary! That’s why we didn’t release the model (blog.openai.com/better-langu…). We’ve anticipated the need to restrict future publishing in our charter (blog.openai.com/openai-chart…) and co-authored a report on specific future threats (blog.openai.com/preparing-fo…)
1
5
2
55
Congrats to @AlecRad, @wuthefwasthat, @jluan, Dario Amodei, @DanielaAmodei, @ilyasut, @rewon, @jackclarkSF, @Miles_Brundage on an amazing and deliberate release.
3
29
This model is the most visceral result so far showing that deep learning isn't just a tool for mapping inputs to human provided labels — there's hope for it doing the kind of self-discovery described in this HN content from earlier today: news.ycombinator.com/item?id…
1
3
36
An OpenAI employee printed out this AI-written sample and posted it by the recycling bin: blog.openai.com/better-langu…
99
654
290
1,963
My first exposure to AI was reading Alan Turing's 1950 paper on the Turing Test, but I became discouraged after realizing we'd made very little scalable progress since then.
Hard to look at these results and still feel that way.
1
2
1
34
Extremely exciting progress in natural language processing, via unsupervised learning on a small fraction of the Internet: nitter.vloup.ch/openai/status/10…. Post contains coherent AI-written paragraphs of text — a leap from the few sentences we could generate two years ago.
We've trained an unsupervised language model that can generate coherent paragraphs and perform rudimentary reading comprehension, machine translation, question answering, and summarization — all without task-specific training: blog.openai.com/better-langu…
2
46
3
206
Really enjoyed my conversation with @reidhoffman onstage at @StartupGrind today, talking about OpenAI and the shape of AGI progress: youtube.com/PGZg750Z-24?t=10864
4
12
2
48
Enjoyed my interview with @emilychangtv today, talking about President Trump’s new executive order on AI: youtube.com/watch?v=q8WWJSSF…
4
14
60
Sundance Film Festival is the only event I've attended where the optimal strategy is to watch as many movies as possible. (Watched 6 movies yesterday!) Favorite part is when you get to watch a premier and it’s the first time the cast has seen it — Q&A afterwards is so heartfelt!
4
2
86
We use different definitions of RL. In mine, any problem can be phrased as a one-step MDP (such as in arxiv.org/abs/1611.01578), and zeroth-order optimization is a special case. Can debate definitions, but I use mine because algos like PPO are doing RL regardless of MDP used.
3
28
