AI Safety

The US government has allowed Anthropic to release its powerful Mythos AI model to select companies and organizations after revising license requirements following an earlier export block due to national security fears. …

The US Commerce Department allowed Anthropic to release its Claude Mythos 5 AI model to a select group of trusted partners, marking a shift from open-access developer platforms to a regulated, government-vetted ecosystem…

Anthropic received US government permission to redeploy its Mythos 5 cybersecurity AI model to over 100 US organizations, including major corporations and government agencies, after suspending access on June 12 due to na…

A developer built mcp-conform, a linter for Model Context Protocol servers that checks for missing tool annotations, injection-prone descriptions, and registry readiness before publication. The tool runs against the live…

A small company, 2Brains Inc., claims to have solved AI hallucinations architecturally, challenging the prevailing industry belief that scaling compute will fix the problem. The company's approach, which separates langua…

A developer lost $12,000 in fraudulent Stripe charges after an AI agent hardcoded an API key into a public GitHub commit. The incident highlights the risk of giving LLMs direct access to codebases via the Model Context P…

OpenAI has delayed the public release of its GPT-5.6 AI model after the U.S. government requested a security review, marking increased regulatory oversight on advanced AI. The model will now be available only to governme…

OpenAI deferred the public rollout of its GPT-5.6 model family, including the Sol variant, at the request of the Trump administration, which sought time for federal security review. The company restricted access to vette…

An engineer warns that LLM judges used to evaluate other models in production pipelines are often unvalidated, leading to silent failures where scores do not correlate with human judgment. The developer proposes treating…

The Trump administration has allowed Anthropic to redeploy its powerful cybersecurity model Mythos 5 to over 100 US government agencies and companies, including non-American employees, after a two-week ban. The decision …

A study led by researchers at the Centre for Addiction and Mental Health (CAMH) in Toronto found that a machine learning model designed to predict aggression on psychiatric wards systematically overestimates risk for Bla…

A developer argues that enterprise AI workflows, particularly in banking, require structured dissent and deterministic rules rather than simply adding more AI agents. The post outlines a multi-agent system where speciali…

A developer scanned 87 MCP servers for security hygiene using a deterministic rule engine, publishing a leaderboard that grades servers from 100 (clean) down based on critical, high, medium, and low findings. The top 24 …

The Commerce Department partially restored Anthropic's Mythos 5 access for about 100 trusted partners and federal agencies after a June 26 letter from Secretary Howard Lutnick revised license requirements, while the publ…

Anthropic is working to restore access to its Claude Fable 5 and Claude Mythos 5 AI models after a U.S. government export control directive suspended access on June 12, 2026, over national security concerns. The company …

Anthropic's Mythos 5 cybersecurity model has been partially reinstated for a select group of organizations after the Trump administration revised license requirements, but the public-facing Fable 5 model remains unavaila…

The Trump administration has eased restrictions on Anthropic's advanced AI model Claude Mythos 5, allowing over 100 US organizations, including corporations and government agencies, to regain access. Commerce Secretary H…

A Forward Deployed Engineer working on AI adoption reports that TDD cannot guarantee design correctness, especially when AI generates tests. The engineer describes an incident where a function signature change broke call…

A developer built self-audit, a zero-dependency CLI tool that checks AI-generated text for completeness, consistency, groundedness, and honesty. The tool, inspired by Anthropic's Constitutional AI framework, catches issu…

NLnet Labs updated its policy on June 26, 2026, to restrict contributions generated by large language models, requiring all code and documentation to be authored by a human and mandating disclosure of any LLM use. Non-co…