PM Keir Starmer sets out sweeping restrictions for under-16s including banned platforms (TikTok, Instagram), expected curfews, and age verification mandates. 'Calling time on a system that's failing o
Anthropic's Frontier Red Team analysed 832 banned accounts found using Claude for malicious cyber activity. Medium-to-high risk actors jumped from 33% to 56% in just 6 months. AI is being used for lat
The US government issued an export control directive blocking access to Anthropic's most advanced models. The directive cites concerns over AI capabilities that could enable sophisticated cyber operat
Tata Consultancy Services will integrate Claude into systems used by banks, insurers, and government agencies, with enterprise-grade security controls.
DXC Technology, a Fortune 500 IT services company, announced an alliance to embed Claude into mission-critical enterprise systems.
A national fellowship program for early-career professionals passionate about extending the benefits of AI safety across American communities.
Anthropic's latest Opus-class model upgrade delivers across coding benchmarks, agentic workflows, and professional use cases with improved reliability for long-running tasks.
Landmark research published with Verizon's 2026 Data Breach Investigations Report shows AI is breaking traditional threat assessment models. The MITRE ATT&CK framework does not capture AI-enabled atta
The AI company has taken a significant step toward going public, filing its confidential S-1 registration statement with the US Securities and Exchange Commission.
Anthropic extends its transparency initiative to approximately 150 new organisations globally, expanding independent access to evaluate model safety.
Archive of cybersecurity news and dark web intelligence reports from the past week.
CyberAware UK's DarkWatch system captured over 16,000 dark web signals this week. 8 new fraud networks targeting World Cup 2026 ticket buyers were detected and flagged.