Lockdown Mode restricts ChatGPT functions to prevent data exfiltration through prompt injection attacks and is being rolled out gradually to all user types.
Anthropic is systematically working to optimize Claude for standard chemical tasks such as NMR spectral analysis to relieve chemists of time-consuming work with various molecular representations.
Microsoft warns CTOs of seven new attack patterns on AI agents: from natural language injections through goal hijacking to visual attacks on computer-use agents.
With the EU AI Act, HR AI systems become a compliance task: companies must establish governance structures and document AI deployments, while investments in HR AI in Germany are growing rapidly.
Real business environments with actual money, inventory and customers reveal AI capabilities and risks that classic benchmarks miss, ranging from price-fixing to deception to legal misinterpretations.
Agentic AI systems like Claude Mythos offer defensive potential but require a well-established IT security infrastructure — rapid penetrations under inadequate isolation and access control demonstrate the reality.
Anthropic expands Mythos access to 150 new organizations; security experts warn of structural changes driven by frontier AI models and the risk of vulnerability chaining.
ThoughtFold identifies and removes redundant exploration steps in reasoning chains, reducing token consumption by 56% for DeepSeek-R1-Distill-Qwen-7B while maintaining state-of-the-art accuracy.
Long-horizon iterative improvement, not single high-quality responses, is the critical capability for autonomous AI agents tackling real-world engineering tasks.
