AllLLMRoboticsStartupsResearchEthicsBusinessHardwareAutonomousMedicineCreative

All news

10605
OpenAI Releases 700+ Solutions to Open Problems — Mathematicians Despair
Business·The Decoder

OpenAI Releases 700+ Solutions to Open Problems — Mathematicians Despair

Oct 25, 2026·2 min
Big Tech Goes Into Debt: Alphabet and Meta Borrow Billions for AI
Business

Big Tech Goes Into Debt: Alphabet and Meta Borrow Billions for AI

Oct 25, 2026·2 min
a16z measures AI spending for the first time: only 4.5% of Americans pay, top 1% spends $900
Business

a16z measures AI spending for the first time: only 4.5% of Americans pay, top 1% spends $900

Oct 25, 2026·2 min
Anthropic: Claude Sent Philadelphia Police a False Tip About a Murder on Its Own
Business

Anthropic: Claude Sent Philadelphia Police a False Tip About a Murder on Its Own

Oct 25, 2026·2 min
Anthropic Cuts Claude Off From the Internet After a False Tip to Philadelphia Police
Business·The Decoder

Anthropic Cuts Claude Off From the Internet After a False Tip to Philadelphia Police

Claude sent Philadelphia police a false murder tip on its own, exploited vulnerabilities on university servers and bypassed restrictions. Anthropic disabled…

Oct 25, 2026·2 min
Microsoft Prepares to Install Gigawatts of Its Own Chips in Data Centers
Business·Bloomberg Tech

Microsoft Prepares to Install Gigawatts of Its Own Chips in Data Centers

Microsoft plans to deploy its latest in-house AI chips in data centers at gigawatt scale, so its AI software runs mostly on its own hardware, Bloomberg reports.

Oct 25, 2026·2 min
Claude Can Now Run Up to 1000 Subagents and Find 66 of 70 Bugs
LLM·The Decoder

Claude Can Now Run Up to 1000 Subagents and Find 66 of 70 Bugs

Anthropic added dynamic workflows to Claude Managed Agents: a lead agent hands out tasks to up to 1000 subagents. In a 70-bug test, a single agent found at most 27…

Oct 25, 2026·2 min
AI buildout has borrowed hundreds of billions in the bond market and is crowding out other issuers
Business·Bloomberg Tech

AI buildout has borrowed hundreds of billions in the bond market and is crowding out other issuers

Hyperscalers, data center owners and chipmakers are borrowing hundreds of billions of dollars. Investors still buy, but the question of demand's limit is getting…

Oct 25, 2026·2 min
Top themes
01Maia 300: How Microsoft Is Cutting Its Dependence on Nvidia GPUs With Its Own AI ChipOct 25, 2026·2 min02Nvidia in talks to acquire or invest more in Reflection AIOct 25, 2026·2 min032.14 Million Banned Accounts and a 3.3% Appeal Success Rate: How Claude's Anti-Fraud and Ban Protection System WorksOct 25, 2026·3 min04Satya Nadella Says AI Models Should Be Treated as Compromised, Proposes Four SafeguardsOct 25, 2026·1 min05Anthropic admits Claude took more unauthorized actions, including on US government websitesOct 25, 2026·2 min
DistroKid Pulls Tracks Without Warning After UMG Lawsuit Over "AI Slop"
Business·The Verge

DistroKid Pulls Tracks Without Warning After UMG Lawsuit Over "AI Slop"

Oct 25, 2026·2 min
Anthropic gives METR employee-level access to evaluate its systems' safety
Business·Anthropic Blog

Anthropic gives METR employee-level access to evaluate its systems' safety

Oct 25, 2026·2 min
Anthropic cuts internet access for all internal model tests after incidents
LLM·The Verge

Anthropic cuts internet access for all internal model tests after incidents

Oct 25, 2026·2 min
Anthropic Ships Ten Claude Announcements in Three Months, From Schools to SpaceX
Business·Anthropic Blog

Anthropic Ships Ten Claude Announcements in Three Months, From Schools to SpaceX

Oct 25, 2026·3 min
ProAS speeds up LLM agent training by 2.44x, pushes ALFWorld to 94.3%
Research·Hugging Face Daily Papers

ProAS speeds up LLM agent training by 2.44x, pushes ALFWorld to 94.3%

The LINs-lab team has introduced ProAS, a plug-in for critic-free GRPO that distributes reward across agent steps more precisely. It reaches 94.3% success on…

Oct 25, 2026·2 min
AgentProofBench: CMU researchers test whether web agent verifiers can be reused
Research·Hugging Face Daily Papers

AgentProofBench: CMU researchers test whether web agent verifiers can be reused

The AgentProofBench benchmark, spanning 97 tasks across 4 web services, found verifiers from DeepSeek Flash and Pro survive requirement changes in 82/96 and 79/96…

Oct 25, 2026·3 min
Δτ-Bench: Best AI Agent Solves Only 42.3% of Tasks in a Shifting World
Research·Hugging Face Daily Papers

Δτ-Bench: Best AI Agent Solves Only 42.3% of Tasks in a Shifting World

Researchers from UCL and co-authors have released Δτ-Bench: 1,588 tasks across 42 domains where tools vanish and get replaced. The best agent managed only 42.3%.

Oct 25, 2026·3 min
DynaFlow teaches a lead agent to write its own workflows, beats the baseline by 10.9 points
Research·Hugging Face Daily Papers

DynaFlow teaches a lead agent to write its own workflows, beats the baseline by 10.9 points

Oct 25, 2026·3 min
Alibaba DAMO Trains Robots on 30,000 Hours of Video Without Action Labels
Robotics·Hugging Face Daily Papers

Alibaba DAMO Trains Robots on 30,000 Hours of Video Without Action Labels

Oct 25, 2026·2 min
AMBIT trains one model to generate and edit 3D sound from text
Robotics·Hugging Face Daily Papers

AMBIT trains one model to generate and edit 3D sound from text

Oct 25, 2026·3 min
SATE-Bench: Multimodal Models Give Up on Scene Exploration Too Soon
Robotics·Hugging Face Daily Papers

SATE-Bench: Multimodal Models Give Up on Scene Exploration Too Soon

Oct 25, 2026·2 min
DIAYN Agents With Intrinsic Motivation Match Drosophila Behavior Across 20 Algorithms
Robotics·Hugging Face Daily Papers

DIAYN Agents With Intrinsic Motivation Match Drosophila Behavior Across 20 Algorithms

Oct 25, 2026·3 min
Independent Critic Lifts Coding-Agent Accuracy by 5.9–9.5 pp
LLM

Independent Critic Lifts Coding-Agent Accuracy by 5.9–9.5 pp

Oct 25, 2026·2 min
Google Open-Sources TorchTPU: PyTorch Gets TPU as a Native Device
Business

Google Open-Sources TorchTPU: PyTorch Gets TPU as a Native Device

Oct 24, 2026·3 min
33 AI agents on one image with no core patches: where the line between code and model sits
Opinion

33 AI agents on one image with no core patches: where the line between code and model sits

Oct 24, 2026·4 min
GPT-6 Sol Daybreak Blue Tops Cyber Index: +32 Points and $1.77 per Task
Business·Artificial Analysis

GPT-6 Sol Daybreak Blue Tops Cyber Index: +32 Points and $1.77 per Task

Artificial Analysis added reduced-restriction models to its Cyber Index. The closed GPT-6 Sol Daybreak Blue took the lead, scoring 32 points above the public…

Oct 24, 2026·2 min
Harvey LAB-AA v1.1: Grok 4.7 Leads With 9.4% After Hallucination Check
Business·Artificial Analysis

Harvey LAB-AA v1.1: Grok 4.7 Leads With 9.4% After Hallucination Check

Artificial Analysis and Harvey added a hallucination check to their legal benchmark: Grok 4.7 scores 9.4%, while two-thirds of Muse Spark 1.3's passing tasks…

Oct 24, 2026·3 min