manaltdan's profile picture. Frontier AI Safety & Governance @GoogleDeepmind

Dan Altman

@manaltdan

Frontier AI Safety & Governance @GoogleDeepmind

Dan Altman reposted

We’re proud to announce that Genie 3 has been named one of @TIME’s Best Inventions of 2025. Genie 3 is our groundbreaking world model capable of generating interactive, playable environments from text or image prompts. Find out more → goo.gle/3KGqiYa

GoogleDeepMind's tweet image. We’re proud to announce that Genie 3 has been named one of @TIME’s Best Inventions of 2025.

Genie 3 is our groundbreaking world model capable of generating interactive, playable environments from text or image prompts.

Find out more → goo.gle/3KGqiYa

The melancholy of entering State of AI Report season 🥹. It's like we're gonna blink and the next AI Index will already be here

🪩The one and only @stateofaireport 2025 is live! 🪩 It’s been a monumental 12 months for AI. Our 8th annual report is the most comprehensive it's ever been, covering what you *need* to know about research, industry, politics, safety and our new usage data. My highlight reel:



Dan Altman reposted

We evaluated Gemini 2.5 Deep Think on FrontierMath. There is no API, so we ran it manually. The results: a new record! We also conducted a more holistic evaluation of its math capabilities. 🧵

EpochAIResearch's tweet image. We evaluated Gemini 2.5 Deep Think on FrontierMath. There is no API, so we ran it manually. The results: a new record!

We also conducted a more holistic evaluation of its math capabilities. 🧵

Dan Altman reposted

How many "digital workers" could OpenAI deploy? We did a back-of-the-envelope estimate: restricting to tasks GPT-5 can do, OpenAI has the compute to run ~7 million digital workers. As AI automates more tasks, such a digital workforce could have big implications. 🧵

EpochAIResearch's tweet image. How many "digital workers" could OpenAI deploy?

We did a back-of-the-envelope estimate: restricting to tasks GPT-5 can do, OpenAI has the compute to run ~7 million digital workers.

As AI automates more tasks, such a digital workforce could have big implications. 🧵

Dan Altman reposted

The point of our work isn't to build an artificial human. The universe is full of questions far more interesting than our own reflection. The point is to create a new kind of mind to help us explore & understand the universe better than we can ourselves.


Dan Altman reposted

Evidence Recent AI Gains are Mostly from Inference-Scaling 🧵 Here's a thread about my latest post on AI scaling... 1/14

tobyordoxford's tweet image. Evidence Recent AI Gains are Mostly from Inference-Scaling
🧵
Here's a thread about my latest post on AI scaling...
1/14

Too many sidebars on web tool UXs these days


Dan Altman reposted

I've spent thousands of dollars testing AI models on real-world tasks & inadvertently learned the nuances of the major model APIs. Here's an unordered list of things I've stumbled upon via TaxCalcBench (testing models' ability to calculate tax returns) about the OpenAI,…


Shadowy forces have been subtly increasing heading font sizes this past decade. It’s gone too far.


Dan Altman reposted

"AI isn't replacing radiologists" good article Expectation: rapid progress in image recognition AI will delete radiology jobs (e.g. as famously predicted by Geoff Hinton now almost a decade ago). Reality: radiology is doing great and is growing. There are a lot of imo naive…

In 2016 Geoffrey Hinton said “we should stop training radiologists now" since AI would soon be better at their jobs. He was right: models have outperformed radiologists on benchmarks for ~a decade. Yet radiology jobs are at record highs, with an average salary of $520k. Why?

deenamousa's tweet image. In 2016 Geoffrey Hinton said “we should stop training radiologists now" since AI would soon be better at their jobs.

He was right: models have outperformed radiologists on benchmarks for ~a decade.

Yet radiology jobs are at record highs, with an average salary of $520k.

Why?


Dan Altman reposted

come hang out tonight!

Looking forward to holding a great event in NYC this evening on Verifying International AI Agreements, a very technically rich topic. We’ll have a keynote from @benharack, followed by discussion with @MauricBaker, @janet_e_egan, @jordanschnyc, and @prpaskov. There are a few spots…

manaltdan's tweet image. Looking forward to holding a great event in NYC this evening on Verifying International AI Agreements, a very technically rich topic. We’ll have a keynote from @benharack, followed by discussion with @MauricBaker, @janet_e_egan, @jordanschnyc, and @prpaskov. There are a few spots…


Loading...

Something went wrong.


Something went wrong.