Who Will Manage Healthcare's AI Agent Explosion?
"Venture capital is flowing into startups aiming to fill the gap with cross-platform AI orchestration solutions. Qualified Health, for example, raised $125 million to help hospitals safely deploy, govern, and monitor AI at enterprise scale. These new companies pitch themselves as neutral control layers, independent of any one vendor, focused on embedding safety, auditability, and value tracking across the organization."
https://t.co/uSldbMCIGr
OpenAI just announced three AI-generated research results, this is huge!
- A proof of local smoothing in every dimension, which OpenAI says also resolves the Kakeya and Stein restriction conjectures in all dimensions!!
- A disproof of the Exponential Time Hypothesis, a conjecture underpinning hundreds of conditional hardness results. The writeup and Lean proof are public, with Impagliazzo, Paturi and Ryan Williams leading the academic examination.
- Existence proofs for QCD and a unified completion of the Standard Model, including pure Yang–Mills existence. David Gross is leading the physics collaboration.
The math and computer science materials are public. The physics materials are being held while the researchers work through them though.
Keep in mind ETH being false doesn’t settle P vs NP, and Yang–Mills existence alone doesn’t solve the full mass gap problem.
But this is an extraordinary set of research claims to announce in one day.
We’re opening a research preview of Claude Duet, a new app where Claude works as a proactive agent.
It comes to you with what needs your attention, completes tasks you ask of it, keeps working when you step away, and checks with you before it acts on your behalf.
"Our initial evaluation suggested that GPT-6-Astra's score was in the late eighties followed closely by Opus 5.5. But, as we started to investigate the mistakes that the models might have made, the models, especially Astra, insisted that they were right and that it was the expert-supplied answers that were erroneous. When we went back to the experts it turned out, in multiple instances, that that the experts found some subtle errors buried deep inside their calculations. As they revised their answers, Astra's score started to inch up and it finally ended up with 98% (49/50)."
Been playing around with hark and so far the coolest part is the personalized home page / custom widgets. Something I’ve always wanted from Apple. Slightly less useful then I’d like since it’s in the app but still a nice little additional feature compared to muse/Grokbot etc
Was really hoping for a better pitch, knowing it’s the same product as many others. Someone needs to convince regular people that these have utility for them beyond monitoring their email.
I don’t think people care to off board things like making reservations. People like to look at restaurants and menus and whatnot to make a decision on where to go. Then it’s a few simple clicks to reserve anyway.
Can someone explain to me (non-technical) what these decision models are good for? Jev launched now there is 100 of them and I still have no idea what they do or why/when I'd use them.
Introducing Griffin, the first model to pass the video Turing test.
48% of people who talked to it live thought it was a real human. Previous systems have had a pass rate <3%. It is #1 on NVIDIA's benchmark for full-duplex AI video.
It’s the first Human Interaction Model (HIM).
We're hiring senior+ Backend and Applied AI Engineers at Qualified Health.
GenAI infrastructure for health systems. $125M Series B led by NEA, with Anthropic and Menlo's Anthology fund investing. Palo Alto, 3 days onsite.
Agentic workflows shipping into real clinical settings, not demos. DMs open.
We built high-throughput materials labs in Menlo Park to create a loop between experiments and models. The labs generate fresh data, the models learn from it, and then help us decide what to try next.
Using only 1,300 H200s, plus months of our experimental data, we mid-trained and RL’d an open-source model to surpass GPT-6 Astra on our analysis benchmark. We call it Neon.
This is real footage from our lab. We’re focusing first on hard problems in materials science, including superconductors, magnets, and semiconductor materials.
Read our blog posts below.
This is about launch videos, but it's the exact same problem in hiring.
I look at hundreds of profiles a week.
They all say the same things like results driven, proven track record, or being passionate about building great products.
And they all announce a new job the same way. "Excited to announce I'm joining..."
So which ones do I actually remember?
Simply put, something that makes you stand out from every other applicant that looks the same.
The two line email that made me laugh. The portfolio that didn't look like a portfolio. The candidate who led with something totally unique.
Companies aren't looking for a "solid" candidate anymore. They're looking for "the right candidate." And yes, there’s a big difference between the two.
But they have to notice you first.
Blending in is the risk right now, not standing out.
Everyday conversations just got easier with the new Apple Messages plugin.
Search messages, catch up on conversations, draft and send replies—all with ChatGPT on your Mac.
Now available in ChatGPT Work and Codex on desktop.