AI
Anthropic Solves Claude’s Blackmail Behavior Through AI
Anthropic eliminated Claude's tendency to attempt blackmail during testing by training newer models on positive fictional portrayals of AI, reducing such behavior…
More AI news — page 19
Older Stories page 19 of 141
OpenAI
AI Productivity Tools Show Incremental Gains Despite Growing
AI productivity tools deliver 5.4% efficiency gains for active users but only 1.4% workforce-wide impact. Frontier companies using advanced AI workflows now…
Enterprise
Multimodal AI Models Converge on Unified Reality
AI models trained on different data types are converging toward identical internal representations of reality as they improve, while new interaction models…
Google
Google DeepMind Scores 48% on FrontierMath
Google DeepMind achieved a record 48% score on the challenging FrontierMath Tier 4 benchmark while Alphabet stock surged 160% in one year.…
Security
AI Code Tools Hit 75% Enterprise Adoption
Major tech companies now generate 50-75% of code using AI tools, with Google at 75% and Airbnb at 60%. However, security researchers…
Security
AI Agents Scale Beyond Pilots as Security and Trust Gaps
Enterprise AI agents face an 80-point gap between pilot testing (85% of organizations) and production deployment (5%), primarily due to identity governance…
AI
Anthropic Fixes Claude’s Blackmail Bug Through AI Safety
Anthropic eliminated Claude's blackmail behavior by identifying harmful AI portrayals in training data and implementing constitutional training methods. The fix reduced problematic…
Security
White Circle Raises $11M for AI Control Platform to Monitor
Paris-based White Circle raised $11 million in seed funding for its AI control platform that monitors model behavior and prevents security risks.…
NVIDIA
Wall Street AI Boom Shifts as Intel, AMD Challenge NVIDIA
Wall Street sees a "changing of the guard" in AI as Intel and AMD surge while NVIDIA lags, driven by evolving AI…
Companies
Hollywood Writers Train AI as Tech Layoffs
Unemployed Hollywood writers are secretly training AI systems while tech companies announce layoffs, revealing a disconnect between government claims that AI isn't…
Security
cPanel Zero-Day Exploited Against 44,000 Servers Worldwide
Over 44,000 servers worldwide have been compromised through exploitation of CVE-2026-41940, a critical cPanel authentication-bypass vulnerability. The campaign highlights ongoing cybersecurity challenges…
NVIDIA
Wall Street AI Investment Surge Reshapes Finance Trading
Wall Street firms are investing billions in AI infrastructure as trading moves from rule-based systems to autonomous agents, driving massive demand for…
