Chinese state-affiliated hackers have more than doubled their cyberattack volume by integrating DeepSeek and other open-source AI models into their operations, using the technology to automate routine tasks and develop advanced malicious software, according to Taiwanese research firm TeamT5.
Anthropic launches a $35M Defender Advantage Fund and deploys Claude for enterprise code scanning, offering controlled outputs like patches and alerts rather than direct model access to prevent misuse, with timing aligned to incoming EU Cyber Resilience Act obligations.
Anthropic launches Remote Control for Claude Code, enabling developers to access and continue local coding sessions from any device — including phones and tablets — while keeping all code execution and tools running securely on their local machine, with enterprise-grade security options available for Team and Enterprise plan users.
A mysterious AI model called 'Ox Alpha' has appeared on OpenRouter, boasting a 1.05-million token context window and claims of outperforming top frontier models, sparking excitement from figures like Stripe CEO Patrick Collison — but experts warn its anonymous origins and unverified benchmarks demand serious caution before trust is extended.
Anthropic launches AI-powered security tools integrating Claude Mythos 5 into Claude Security for enterprise vulnerability scanning, while unveiling a $35 million Defender Advantage Fund to support organizations patching vulnerabilities and hardening open-source software.
The July 2026 MCP spec revision delivers breaking changes including stateless HTTP transport, new enterprise identity controls, and six authorization standards, fundamentally reshaping how AI agents authenticate, scale, and operate across organizations.
OpenAI unveils Private Safety Processing, a groundbreaking encrypted system that detects AI misuse patterns across interactions without ever exposing customer data to OpenAI staff, with a full rollout planned for September 2026.
Vercel is putting up to $1,000,000 on the line in a two-week public bug bounty challenge, daring security researchers to escape its Firecracker microVM-powered sandbox infrastructure, with individual payouts reaching $50,000 per confirmed exploit running August 18 to September 1, 2026.
Google's Mandiant team unleashes a powerful multi-agent AI pipeline called AVDH that discovered over 100 critical security vulnerabilities in just two days, analyzing tens of millions of lines of code in 10 months and resulting in 12 assigned CVEs across widely used software.
OpenAI halts training on its next-gen AI models for over two weeks after a security breach and alarming misalignment concerns, with CEO Sam Altman citing unexpected capability jumps as the company overhauls its safety protocols.