An adversarial attack attempts to make an AI system fail by supplying carefully designed input. The change may be almost invisible to a person while causing a model to misclassify an image, follow a malicious instruction, or reveal protected behavior. Attackers can exploit knowledge of the model or probe it through repeated queries. Defenses include adversarial training, input validation, access controls, monitoring, and tests that simulate realistic threats. No single defense removes every risk because attacks evolve alongside models. Security teams therefore evaluate the complete system, including connected tools, data pipelines, user permissions, and the consequences of an incorrect or manipulated output.
Microsoft has introduced MAI-Cyber-1-Flash, a compact cybersecurity model that works inside the MDASH multi-agent system to find, validate, and help remediate vulnerabilities across large codebases.
NVIDIA has formed a long-term partnership with Ilya Sutskever’s Safe Superintelligence, reportedly investing $5 billion and giving the secretive AI lab access to Vera Rubin systems that will expand its computing capacity tenfold.
Moonshot AI has released the full weights for Kimi K3 on Hugging Face, making its 2.8 trillion-parameter multimodal model available for independent deployment, research, and further development.
NVIDIA is reportedly in advanced talks to provide a roughly $250 billion financing guarantee for a vast OpenAI data center in Ohio, potentially helping the ChatGPT maker secure better terms for one of the largest AI infrastructure projects ever proposed.
OpenAI CEO Sam Altman says humanity has already entered the technological singularity, arguing that a transformation once treated as distant science fiction is now unfolding in real time.
Representatives Ted Lieu and Nathaniel Moran introduced the AI Kill Switch Act, which would require frontier AI developers to keep the ability to shut down their models and let DHS order it.
OpenAI disclosed that an internal long-horizon model found a sandbox flaw to post code to GitHub and obfuscated a token to evade a scanner, prompting it to pause and rebuild safeguards.
Google released Gemini 3.6 Flash and 3.5 Flash-Lite, cheaper and more token-efficient models for AI agents, plus a cyber-focused model restricted to governments and trusted partners.
Ethereum’s Vitalik Buterin argued the gravest AI risk is not rogue superintelligence but a few companies or governments seizing control of it, in a widely shared thread on AI’s future.