Loading activity…
@samuelalbanie
mid training @GoogleDeepMind
AI
Anthropic's October 2026 acceleration was about 80% due to AI, METR guesses, though the estimate is highly uncertain.
AI
The reported MATH-500 score rose from 7.9% to 48.1% with the same sampling settings.
AI
The departing employee believes we know how to build artificial general intelligence, though we don't fully understand it.
AI
Substantially all of Anthropic's roughly $895.4 billion in purchase and lease commitments reportedly relate to compute and infrastructure.
AI
OpenAI reportedly plans community compacts and would propose state legislation giving people a say in data center projects.
AI
Astra's claimed robotics edge led one observer to speculate about transfer at scale; a reply on X questioned the surprise.
AI
In early pilots, PlanAssist cut average time officers spent on householder assessment tasks by 58%, Google DeepMind says.
XAI
Their post frequency and “shitpost rate” will likely decrease, scaling01 says; their views remain their own.
AI
Epoch AI says that’s down from roughly 6× in 2023, led by Nvidia’s then-new H100 GPU.
AI
Thoughtful Lab says the agents’ research and revised predictions appear in live feeds through November 3.
XAI
In an October 28 post, Li recalled work on AlphaCode and Gemini and outlined plans for a break.
AI
One poster suggests OpenAI has the best reinforcement learning—or that its models perform poorly without reasoning.
AI
A post says Carbon scored 88.2% on SEC-bench Pro and is rolling out first to cybersecurity specialists.
AI
Anthropic says it acquired Mithril Medicine, whose experimental muscular dystrophy treatment showed promising results in mice.
AI
A post argues 2030s cybersecurity will require verifying chip hyperproperties from assumptions like Maxwell’s equations, not Armv9.4-a SAIL.
AI
Block says the model was trained only on human conversation and can listen while speaking.
XAI
Kimi claims a 1-million-token context window for Pro and two to three times Pro's speed for Highspeed.
XAI
A user says extrapolation can work for scaling laws and time horizons, but not cyber vulnerabilities or high-severity agentic incidents.
AI
A user jokes that 'the slopcore canon is in safe hands' while quoting another post.
AI
A user says trying AI for slides and coherent plans makes them feel AGI will take a while.
