security teams

Anthropic is broadening access to its Claude AI system for vetted cybersecurity teams, in a move reported by The Hacker News. The announcement comes alongside figures from Glasswing, which says its use of the technology has uncovered 129,000 software flaws. The rollout signals Anthropic's effort to position Claude as a defensive security tool while keeping it away from malicious actors through screening of approved teams. Coverage is drawing attention from security professionals weighing AI's growing role in vulnerability discovery.

A new open-source tool called AI Agent Gateway has been released, designed to keep credentials out of AI agent configurations. Instead of embedding API keys and secrets directly in agent setups, the gateway manages authentication centrally, reducing the risk of leaked credentials. Coverage highlights it as a practical security step for teams deploying AI agents that need access to multiple external services.

A tool called PQC Triage has been released that lets developers paste a software dependency manifest and see which cryptographic algorithms in their stack would be broken first by a quantum computer. It aims to help teams prioritise migration to post-quantum cryptography, with discussion among security and open-source developers about practical quantum-readiness.

IBM and Red Hat have released fixes for more than 400 security vulnerabilities affecting open source software components. The patch batch addresses flaws across Red Hat's enterprise Linux offerings and associated packages, and administrators are being urged to apply the updates promptly to reduce exposure to potential exploits.

Anthropic's frontier red-teaming report claims a popular Chinese open-weight AI model has so-called Mythos-class hacking abilities, with weak safeguards on openly available weights. The finding is fueling debate over security risks of freely downloadable frontier-scale models and how AI companies assess cyber capabilities of systems developed outside their control.

Security teams are weighing which vulnerability findings should be handled by automation, which by AI agents, and which still require human analysts. The discussion, highlighted by Help Net Security, reflects a broader industry debate over how to divide alert triage work as AI tools become more capable and security teams face growing volumes of findings and limited staff.

Temporal, the fast-growing workflow orchestration company, has hired the team from New York startup Oso to build security controls for AI agents. The acqui-hire brings Oso's authorization expertise in-house as Temporal moves to address emerging security concerns around autonomous AI systems. The move underscores how infrastructure companies are racing to supply guardrails as enterprises deploy AI agents more widely.

Max Verstappen has publicly expressed uncertainty about his Red Bull situation shortly after securing what is described as a breakthrough Formula 1 win. The remarks have drawn attention across the motorsport world, with fans and pundits speculating about what the comments could mean for the four-time champion's future with the team.

Pitcher Michael King and infielder Jake Cronenworth spoke after their team's 4-3 victory, breaking down the performance that secured the narrow win. The two players reflected on key moments in the game in comments carried by MLB.com. No further details about the opponent, date or standout plays are available.