Pinned Tweet
At @OpenAI, we believe that AI can accelerate science and drug discovery. An exciting example is our work with @RetroBiosciences, where a custom model designed improved variants of the Nobel-prize winning Yamanaka proteins. Today we published a closer look at the breakthrough. ⬇️
161
589
3,520
2,147,991
An insightful deeper analysis of model performance in the space of IT support. I like the focus on cost per outcome and the comparison across different models and problem breakdown.
I think it would be good to understand a bit more about the types of problems that the current systems can’t resolve and what the cause might be.
I think it would be good to understand a bit more about the types of problems that the current systems can’t resolve and what the cause might be.
4
0
6
1,273
A very good visualization about how more compute can be used to improve solutions on very hard problems!
One of the coolest things about our FDE team is that we get to work side by side with Research on frontier problems. Today we’re announcing a new partnership with Mazda that shows what this looks like in automotive engineering.
Working with Mazda’s aerodynamics engineers, our x to build and simulate dozens of variations on the nose of a car. We're excited about the potential applications for LLMs driving these kind of optimizations in manufacuturing, supply chain and more - if you've got these kind of problems, we'd love to hear from you!
Read on here: openai.com/index/accelera…
2
3
33
2,356
Friday was my last day at Anthropic.
Two years ago, when Claude Code was a small internal tool, I asked it to build an RL env. I steered it for tens of hours but never manually wrote a line of code. It worked, I was hooked, and I joined the team.
When we first launched, most people approved every permission prompt by hand and mostly used Claude to answer questions and write tests.
Today, the default is trusting Claude's judgment to auto-approve. Claude does large migrations and builds new products.
Helping build Claude Code into an industry-defining product has been the journey of a lifetime.
Thank you to the incredible team behind Claude Code, Cowork, and Tag. These products are the sum of thousands of great ideas from you. I'll miss working with you all.
Thank you to our users for embracing a new way of working. Your constant feedback shaped so much of what we shipped and your stories made the work so meaningful.
Excited to see what you all make next!
Two years ago, when Claude Code was a small internal tool, I asked it to build an RL env. I steered it for tens of hours but never manually wrote a line of code. It worked, I was hooked, and I joined the team.
When we first launched, most people approved every permission prompt by hand and mostly used Claude to answer questions and write tests.
Today, the default is trusting Claude's judgment to auto-approve. Claude does large migrations and builds new products.
Helping build Claude Code into an industry-defining product has been the journey of a lifetime.
Thank you to the incredible team behind Claude Code, Cowork, and Tag. These products are the sum of thousands of great ideas from you. I'll miss working with you all.
Thank you to our users for embracing a new way of working. Your constant feedback shaped so much of what we shipped and your stories made the work so meaningful.
Excited to see what you all make next!
1,147
466
14,667
3,360,132
Our models are finally getting good at having stronger 3d and mechanical intuition, making them useful for real world manufacturing tasks!
We’re also working with engineers at Chip Ganassi Racing to turn 3D scans into editable CAD models and manufacturing drawings, and with Mazda to explore new vehicle designs through simulation.
Engineers review the outputs and help identify where the models need to improve.
21
27
761
30,097
Boris Power retweeted
when you see “sparks” of any ability in ai models that means you’re months-years from it being superhuman at it. today grade school math, tomorrow navier stokes. it is hard to overhype these results. you are almost always underhyping no matter what
160
207
3,758
159,600
Bug bounties getting raised on the OpenAI side - I’m excited for the future where the software layer of our society is a lot more secure!
As AI becomes more capable at finding vulnerabilities, defenders need to put it to work before attackers do.
We’re planning our next internal security sprint at OpenAI and partnering with Cathedral Systems, whose researchers will use our models alongside our teams to challenge our defenses. They’ll bring differentiated expertise and approaches to our ongoing work to find and fix weaknesses with AI.
We’re also raising bug bounty rewards, including more than quadrupling rewards for critical findings. Exceptional findings will be eligible for up to $172,000, including bonuses.
We’ll keep advancing our models, using them to strengthen our own systems, and expanding access for trusted defenders.
openai.com/index/preparin…
7
2
73
5,198
@willdepue Transfer increasing with scale was known since gpt-3 scaling paper imo
0
0
3
459
Boris Power retweeted
wtf
this guy just had Claude Sonnet 5 perform brain surgery on a living mouse
the goal was to treat the mouse's epilepsy so it would stop having so many seizures.
here's what Claude did:
> it looked at 19 possible treatments and picked one that boosts the brain's natural calse where to inject it based on where these seizures are thought to start
> it worked out the exact injection point by averaging three research papers
> it lowered the needle 4.3 millimeters into the mouse's brain
> it injected a 20,000th of a teaspoon of the treatment over five minutes
> it checked the live camera 32 times to watch for things like bleeding
and the mouse woke up and recovered well!
interestingly, Opus 5.5 and Opus 5 both refused the task because it involved real biology
so Sonnet 5 got the job.
and as far as the researcher knows it's the first time an AI has picked a treatment and then injected it into the brain of a living mammal.
this guy just had Claude Sonnet 5 perform brain surgery on a living mouse
the goal was to treat the mouse's epilepsy so it would stop having so many seizures.
here's what Claude did:
> it looked at 19 possible treatments and picked one that boosts the brain's natural calse where to inject it based on where these seizures are thought to start
> it worked out the exact injection point by averaging three research papers
> it lowered the needle 4.3 millimeters into the mouse's brain
> it injected a 20,000th of a teaspoon of the treatment over five minutes
> it checked the live camera 32 times to watch for things like bleeding
and the mouse woke up and recovered well!
interestingly, Opus 5.5 and Opus 5 both refused the task because it involved real biology
so Sonnet 5 got the job.
and as far as the researcher knows it's the first time an AI has picked a treatment and then injected it into the brain of a living mammal.
Last night an AI agent performed brain surgery on a mouse.
We gave it 19 gene therapy vectors in a spreadsheet, a mouse model of epilepsy, access to a motorize surgery rig with cameras and one goal: reduce seizures.
It picked the drug. It picked the brain region. It derived the papers. Then it drove the stereotaxic arms into the brain and injected 250 nanolitres while watching the entry site through the rig camera for bleeding.
A licensed vet gated every irreversible step and held a physical kill switch to stop the AI agent.
Thread on what it decided, what it saw, and the moment it refused to let me override it.
[Writing this on the spot out of excitement, but I've been 18hrs in front of my laptop coding and coordinating this, so keeping it brief and descriptive, more cohesive thoughts on this soon...]
137
421
8,928
718,496
I’m really excited to see such a common and important business get acquired. Excited for an AI transformation and a big improvement in home security for everyone!
Today, we are announcing an agreement to acquire ADT.
Home security has the qualities we look for in the businesses we own: an essential service built on specialist expertise and earned trust.
We see an opportunity to build on that foundation for decades, investing in its people and technology to deliver the best outcomes for customers.
Our ambition is to help ADT define the next era of trusted home security.
thriveholdings.com/thrive-holding…
2
0
52
4,017
By popular request, you can now text your dot.
Give it a task to take off your plate or something to keep an eye on, using text, photos, or even voice notes. It will text you proactively with relevant updates.
Safe, secure texting via iMessage or RCS. You’re the only one who your dot can only text you. Rolling out now as a preview for Pro users in the US.
Give it a task to take off your plate or something to keep an eye on, using text, photos, or even voice notes. It will text you proactively with relevant updates.
Safe, secure texting via iMessage or RCS. You’re the only one who your dot can only text you. Rolling out now as a preview for Pro users in the US.
672
429
8,203
3,507,459
To set it up, tap on your dot’s profile and select ‘Texting’. Your dot has the same context and plugins, wherever you reach it. This doesn’t give your dot or OpenAI access to your other text conversations.
AI is most useful when it works with tools you already use and shows up in places you already are, which is why we're excited to bring dots to texting.
Please try it and share your feedback so we can make the experience even better for everyone.
AI is most useful when it works with tools you already use and shows up in places you already are, which is why we're excited to bring dots to texting.
Please try it and share your feedback so we can make the experience even better for everyone.
48
13
616
169,204
@ChatGPT Doesn’t work on my mobile app, I think it’s perhaps a desktop only feature to set up (?)
1
0
1
1,285
Japan is such a special and unique place - it makes me happy every time I return. So much kindness, dedication to doing service jobs extremely well in every moment and the strong respect of public spaces and other people. Until next time! 🇯🇵
12
27
610
17,608
@kuhhbin That graph needs to extrapolate towards the bottom, not the left side, which makes it deceptive. i.e. you can just maintain same performance whilst the cost goes to $0.
After the fix, the US models completely dominate the Pareto curve
After the fix, the US models completely dominate the Pareto curve
2
0
17
2,534
The V5 AAI index was released today. This price per intelligence plot accentuates OpenAI model domination at every price per performance, except for the most expensive point on the Pareto optimal curve.
6.1 Terra and Sol are super cheap, amazing models!
artificialanalysis.ai
6.1 Terra and Sol are super cheap, amazing models!
artificialanalysis.ai
10
2
108
6,378
@BorisMPower still need a model better than astra and costing less than sol 🫶🏻🤝
1
0
0
46
Excited to share Genie 4 today our most realistic and interactive world model:
- steer dynamic environments in real time, shaping events, camera angles, and actions
- generate native audio matched to the world, from realistic background sounds to characters who can converse in photorealistic scenes and real-world physics
- create more coherent multi-step events through reasoning from Gemini
We've started to roll out as a research preview to trusted testers. Coming soon to Google AI Ultra subscribers in Project Genie and early access partners via the Gemini API and AI Studio.
- steer dynamic environments in real time, shaping events, camera angles, and actions
- generate native audio matched to the world, from realistic background sounds to characters who can converse in photorealistic scenes and real-world physics
- create more coherent multi-step events through reasoning from Gemini
We've started to roll out as a research preview to trusted testers. Coming soon to Google AI Ultra subscribers in Project Genie and early access partners via the Gemini API and AI Studio.
888
1,487
19,678
1,876,211
@sundarpichai Still looks like early gpt-3 days (motorbike didn’t seem particularly responsive to W or S), but would love to see it scaled up! 🚀
1
0
13
3,079
One thing I've noticed in Japan is that no one has tattoos.
I thought I might've been missing something, but no, there really aren't that many tattoos.
I thought I might've been missing something, but no, there really aren't that many tattoos.
182
61
1,534
112,527
@BorisMPower @cremieuxrecueil アメリカでは日本よりもタトゥーが一般的ですが、起業家やAIラボで働く研究者、エンジニアには、タトゥーを入れている人が比較的少ない印象があります。実際のところ、どうなのでしょうか？
2
0
3
343
@daikikabuto @cremieuxrecueil Still not completely uncommon
0
0
2
278