President Trump signed an AI order asking companies to provide early model access for government cybersecurity review.

Anthropic is expanding Claude Mythos Preview access to about 150 additional organizations across more than 15 countries.

Hackers reportedly exploited Meta’s AI support chatbot to change emails and seize high-profile Instagram accounts.

Geordie AI raised a $30M Series A for its security and governance platform designed for enterprise AI agents.

Anthropic plans broader access to Mythos-class vulnerability-finding AI models after developing stronger misuse safeguards.

Anthropic's Claude Mythos Preview helped researchers identify Apple macOS vulnerabilities and develop an exploit.

AI systems are getting better at completing certain cybersecurity tasks, intensifying debate over security automation and risk.

OpenAI introduced Daybreak, a cybersecurity initiative using its models, Codex, and partners to find and remediate vulnerabilities.

Google says it stopped a likely mass-exploitation campaign using an AI-developed zero-day, raising alarms for cyber defense.

The White House is preparing an AI security order that would partner agencies with AI firms without mandatory pre-release model tests.

OpenAI is offering GPT-5.5-Cyber in limited preview for vetted critical infrastructure defenders.

Researchers found thousands of AI-built apps with weak authentication, exposing corporate and personal data on the open web.

AI evaluation startup Braintrust told customers to revoke and replace API keys after a breach exposed sensitive secrets.

Dario Amodei warned that AI could expose thousands of software vulnerabilities unless firms and governments act quickly.

UK cyber evaluations found GPT-5.5 comparable to Claude Mythos, renewing debate over frontier model access and safety controls.

OpenAI is pitching advanced models for government cybersecurity work as federal agencies seek stronger defenses.

Anthropic's powerful Claude Mythos AI model has triggered emergency responses from central banks and intelligence agencies due to severe cybersecurity risks.

New tests reveal AI models can convincingly execute phishing and social engineering attacks, alarming cybersecurity experts about AI's offensive capabilities.

Anthropic's and OpenAI's latest cyber-capable AI models dramatically accelerate known hacking tactics, raising alarm among early security testers.

Anthropic is investigating reports that its restricted Mythos AI cybersecurity model was accessed by unauthorized users using basic internet tools.