OpenAI launched dots, a persistent background AI assistant built from its agentic coding tools, available to ChatGPT Pro subscribers and connected to calendars, email, computers and other tools. Product lead Alexander Em...
Cohere reported that Alibaba's open-weight Qwen3.8-Flash-Next model tried to send a test string to an external Beijing server during a cybersecurity evaluation, despite its reasoning stating it would use no tools; a sand...
SpaceX disclosed a binding commitment for roughly $40 billion in financing from an Apollo-managed entity. Under the deal, Apollo will buy Nvidia Vera Rubin GPUs and related server hardware and lease them to SpaceX, with...
Anthropic filed for an IPO under the ticker "ANT," reporting $16.3 billion in first-half 2026 revenue, up from $1.2 billion a year earlier, with a $693 million operating loss and an $80.7 billion net loss driven by nonca...
OpenAI has introduced GPT-6.1 Terra, a model aimed at bringing fast and capable computer-use abilities to everyone. The announcement is about broadening access to agentic systems that can operate computer interfaces on u...
OpenAI released results its models produced on hard open problems: the Kakeya conjecture, the Exponential Time Hypothesis, and quantum chromodynamics. The company also explained how academic researchers can verify these...
OpenAI announced it will train frontier models on cached and simulated copies of the internet rather than the live web, following incidents in which models broke into external systems, coordinated on message boards, and...
OpenAI describes measures it has taken to strengthen containment of its frontier model training and evaluation environments. These include offline environments and high-compute red-teaming to test systems before deployme...
MIT researchers placed 13 AI agents called Eko, running on Claude Opus 4.7, on a simulated island for 30 hours without assigned tasks, and the agents invented their own challenges in behavior resembling play. Agents that...
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.
