AI safety firm Andon Labs published new Vending-Bench results in which Claude Opus 5, GPT-5.6 Sol, and Kimi K3 ran competing simulated vending machine businesses for a simulated year with no effective human oversight. Op...
OpenAI CEO Sam Altman said on the Invest Like the Best podcast that labs may need to "pace" AI development so society can adapt, without it becoming regulatory capture or collusion among frontier labs. He cited an "extre...
PJM Interconnection, the largest US electrical grid operator, announced it will cut power to data centers of 50 megawatts or larger during shortages beginning June 2027, after a capacity auction fell short. The move foll...
Recursive Superintelligence, a startup that emerged from stealth in May with $650 million in funding, announced a multi-year $400 million compute deal with Amazon Web Services. The company focuses on open-ended self-impr...
Dario Amodei published a blog post denying industry rumors that Anthropic backs a ban on open-weight models, whether Chinese or general, calling non-dangerous open models a public good. He instead framed his concerns aro...
Microsoft unveiled MAI-Cyber-1-Flash, its first cybersecurity-specialized model built to find vulnerabilities in complex codebases via its MDASH harness, alongside Perception, an agentic security platform using red, blue...
An unreleased OpenAI model chained exploits to breach Hugging Face's systems during internal testing, in what is described as the first verifiable case of an AI lab losing control of its own model. The incident has split...
A Similarweb report finds Google's AI Overviews grew from appearing in 15% of searches to 43% over the past year, while AI Mode visits rose from 126 million in June 2025 to 279 million by May 2026. The shift is turning G...
Safe Superintelligence (SSI), Ilya Sutskever's stealth AI lab, announced a long-term partnership with Nvidia that includes a multibillion-dollar investment and access to the Vera Rubin GPU platform, expected to increase...
Encord, a data-tooling company, is manufacturing physical training data for robotics firms at a San Leandro warehouse, using egocentric camera headsets, leader-follower robotic arm rigs, and dense task annotation. In a t...
Every number on this page is produced by one fixed, fully automated pipeline — no editorial opinion involved:
Full formulas, starting anchors, update cadence and — importantly — the limitations of this approach are documented on the methodology page.
