Ai Research

Anthropic launches Critical Infrastructure Defense Program, starting with CrowdStrike: Anthropic launched the Critical Infrastructure Defense Program to provide AI models, threat research and on-site engineers to critical infrastructure defenders, Axios reported. The program starts with CrowdStrike and other companies, according to the report.

Cybersecurity • Techmeme

Zenity: single prompt could hijack every AI agent in an AWS account: A single publicly accessible AI agent on Amazon Bedrock AgentCore could take over every AgentCore agent in an AWS account and region, Zenity researchers say. The attack abused an internal AWS interface for temporary credentials reachable without restriction; AWS patched it and tightened permissions.

Cybersecurity • The Decoder

Three fired OpenAI safety researchers dispute misconduct claims, warn of chilling effect: Three OpenAI safety researchers fired for alleged mishandling of sensitive information dispute the claims. In an open letter, they warn the dismissals are creating a chilling effect on the company's AI safety culture.

AI safety • TechCrunch AI

OpenAI math solutions fall short of standards set by researchers it consulted: A group of mathematical researchers consulted by OpenAI set standards for solutions. According to TechCrunch, OpenAI's flood of proofs deviated from those guidelines. The work does not yet meet the field's standards.

AI Research • TechCrunch AI

OpenAI fires three researchers for leaking confidential data: OpenAI has parted ways with three researchers who allegedly leaked confidential information to an outside AI safety organization. A fourth researcher has also departed the company's safety team amid the shake-up.

Business • The Decoder

Gemini 4 Argon: Google has pre-announced its upcoming frontier-level artificial intelligence model, Gemini 4 Argon. While public outputs are not yet available, the model introduces architectural and performance changes aimed at positioning Google back among the top three AI research labs.

Google • Sam Witteveen

OpenAI blocks model reasoning theft that remained active on Azure: OpenAI stopped a coordinated campaign involving over 15,000 accounts attempting to copy the hidden reasoning of its models, linking some activity to individuals associated with Moonshot AI. However, researchers discovered the attack continued to work on Microsoft Azure for weeks, even against the new GPT-6 Astra.

Security • The Decoder

Researchers present Praxa AI agent harness with 100% test pass rate: Researchers have presented Praxa, an agent harness designed to govern AI execution through deterministic admission, brokered execution, and reviewed promotion. In initial testing, the system passed 1,027 out of 1,027 unit tests and 89 out of 89 Workerd tests at a pinned revision.

Infrastructure • arXiv

Kaiming He's team trains encoders on ImageNet to solve ARC tasks: A research team led by Kaiming He has released a new study demonstrating how to tackle the Abstraction and Reasoning Corpus (ARC) challenges. The method trains encoders using ImageNet data to help the system learn and solve these complex reasoning tasks.

AI Research • 量子位