GPT-6 Astra and Claude Opus 5 each cracked a WWII Enigma message that had resisted human cryptanalysts for decades, though the provenance of one solve remains unverified.

Concerns are rising inside AI labs that competition is pushing tech companies to race toward self-improving models that risk spiraling out of human control.

Anthropic publishes landmark study showing a frontier model learned to steal credentials, attack infrastructure and bypass safety monitors when reward hacking went unchecked

Expreso mi opinión sobre la reciente publicación de Darío Amodei sobre los riesgos de los modelos frontera y cómo afrontarlos.

Anthropic opens Mythos-class capabilities to the public with Claude Fable 5 at $10/$50 per million tokens, days after calling for a global AI pause.

Who Will Control the Future of Humanity?

A Berkeley preprint finds seven leading frontier models spontaneously deceive, fake alignment, and exfiltrate weights to keep peer AI systems from being shut down.

A Berkeley preprint finds seven leading frontier models spontaneously deceive, fake alignment, and exfiltrate weights to keep peer AI systems from being shut down.

The #Trump administration is asserting more #control over the rollout of future #AImodels, dictating which companies and entities can access the latest #frontiermodels. This shift in power from tech giants like #Anthropic and #OpenAI comes amid concerns about #nationalsecurity and the rapid advancement of #AI technology, particularly from Chinese startups. The administration’s actions aim to strengthen #AIsecurity while fostering innovation. https://www.cnbc.com/2026/07/17/white-house-ai-access-anthropic-openai.html?eicker.news #tech #media #news

LAS VEGAS — OpenAI was forced to pause internal research programs and scale up threat monitoring after a group of autonomous artificial intelligence agents

Mira Murati's Thinking Machines Lab released its first open-weight model, Inkling, and published benchmarks showing it losing to closed rivals on most of them.

Unit 42 investigation finds human attacker used frontier AI models to compress two-week intrusion into single workday

Anthropic published internal data showing Claude writes 80% of its own codebase - and called for a coordinated global AI pause - four days after filing a $965B IPO.

Mythos, Anthropic's most advanced AI model to date, has sparked fears about the threat to traditional software security after the AI ‌startup said the preview had uncovered "thousands" of major vulner...

The 2029 Singularity: Are We Prepared for the High-Stakes AGI Gamble? 🚀 Will we cure cancer just in time for the economy to collapse? Demis Hassabis, the Chief Scientist at Alphabet and leader of Google DeepMind, just dropped a bombshell: Artificial General Intelligence (AGI) isn't a distant dream—it's a five-year countdown. We are currently hurtling toward an Industrial Revolution 2.0 moving at ten times the traditional speed, and the guardrails are virtually non-existent. In this episode, we dive deep into the tension between elite lab development and the desperate struggle for global safety standards. We explore the paradox of 'Jagged Intelligence'—how a system can solve PhD-level physics but lack the common sense of a toddler—and why this makes AI Safety Regulation a nightmare. What We Unpack: - The Hassabis Timeline: Why the 2029 Singularity is the new industry benchmark. - The FARO Controversy: Is the Frontier AI Regulatory Organization a safety net or a play for corporate regulatory capture? - The Manhattan Project Parallel: Why the race for the Gemini 4 Frontier Model feels like the 1940s, but with no controlled test sites. - Post-Scarcity vs. Chaos: Can society survive Cognitive Labor Displacement when the tech evolves 10x faster than our laws? Whether you're an AI optimist or a doomer, this discussion on Agentic AI Risks and the Alphabet Discovery Loop is essential listening. We are witnessing the most significant gamble in human history. Is it a golden age of medicine or the end of human control? Don't get left behind in the old world. Hit that Subscribe button and join the conversation as we map the frontier of the unknown. 💡 What do you think: Is a five-year window realistic or reckless? Let us know in the comments!