Latest coverage for Artificial Intelligence

Reports claim Anthropic engineers are helping the NSA use its restricted AI model Mythos, known for advanced cybersecurity capabilities. This week, the Financial Times reported that Anthropic has placed approximately six “forward-deployed” engineers inside the National Security Agency to help the intelligence agency use Mythos, its most capable cyber model, for offensive operations. Two people […]

OpenAI has begun rolling out a new Lockdown Mode to ChatGPT for eligible personal accounts to reduce the risk of data exfiltration arising from prompt injection attacks

A researcher has reverse-engineered the iOS SDK that Bright Data embeds in consumer apps and documented how it turns devices, including always-on smart TVs, into exit nodes that relay web-scraping traffic for a data business Bright Data markets heavily to the AI industry

Two things landed within days of each other this week. A security startup reported 21 previously unknown vulnerabilities in FFmpeg, the media library inside almost everything that touches video, all of them found by an autonomous AI agent

Experts: If Implemented Effectively, Directive Could Help Health OrganizationsAlthough President Donald Trump's executive order this week on artificial intelligence barely mentions healthcare, some experts said the directive could potentially have a positive impact on the critical infrastructure sector - if implemented effectively.

House Draft Targets Frontier Labs With Audits While Blocking State AI LawsA House discussion draft would impose safety mandates and independent audits on the country's most powerful artificial intelligence developers, but the fight over a three-year block on state AI laws is already dominating the response.

Microsoft Threat Intelligence identified a prompt injection pathway in Claude Code GitHub Action that allowed access to workflow secrets under specific conditions. This research examines the attack chain, responsible disclosure process, Anthropic's mitigation, and guidance for securing AI-powered CI/CD workflows. The post Securing CI/CD in an agentic world: Claude Code Github action case appeared first on Microsoft Security Blog.

Bob Costello on Voluntary Plan's Impact on Collaboration - and CISA's Pivotal RoleFormer CISA CIO Bob Costello said President Trump's voluntary AI cybersecurity review order provides a workable foundation for government-industry collaboration, though agencies will need time and resources to meet accelerated 30-day evaluations of advanced AI systems.

AI worms, or "viruses with wings and brains," adapt to new environments, seek out vulnerabilities, and will likely strike within a year, researchers say.

Conference Highlights AI Maturity, Agentic Risks and Human Factors in CybersecurityISMG editors reflect on key themes from Infosecurity Europe 2026, including AI's role from buzzword to business strategy, the risks of agentic systems in critical infrastructure and why human-to-human trust is emerging as a defining factor in cybersecurity.

Phishing, shadow AI, malicious extensions, and credential theft increasingly happen inside the browser. Keep Aware explains what the 2026 Verizon DBIR reveals about browser-layer security gaps and modern attacks. [...]

The White House's executive order establishes voluntary framework for early government access to frontier models while investing in federal security.

Lloyds Banking Group shared its approach for securing agentic AI workflows, with a mix of hands on experimentation and cross functional governance

Eighteen months ago, the AI SOC was a marketing line. Today it's a budget item. The category has crossed over from interesting to inevitable, with billions of dollars now flowing into AI-powered security operations platforms, agentic SOC tools, and AI co-pilots built into every layer of the security stack. The data shows SOCs are buying, deploying, and standing up AI capabilities at the fastest

The OWASP agentic AI security framework helps organizations assess governance maturity vs adoption and adjust governance as needed

Ox Security field CTO, Boaz Barzel, makes the case for vibe security to tackle AI agent coding risks

Frontier AI Models Accelerate Tasks Once Measured in Months to Weeks or DaysSentinelOne will cut about 240 employees, citing productivity gains from frontier AI models that have dramatically accelerated internal workflows, while redirecting savings into AI security, cloud, data and endpoint initiatives to drive long-term growth and profitability.

Brave has announced the public release of Brave Origin, a paid minimalist version of its browser that strips out cryptocurrency, AI, rewards, and other monetization-focused features. [...]

Both Parties Urge Judges Not to Throw Out LawsuitThe U.S. Department of Defense denied a request by AI giant Anthropic to reconsider its blacklisting on national security grounds, federal attorneys disclosed in a Thursday court filing. That may come as news to Anthropic, which said it "hopes the secretary will rescind his designation."

A surge in real-world attacks against agentic AI systems is reshaping how we think about risk. Based on 12 months of red teaming, this update introduces seven new failure modes, from supply chain compromise to goal hijacking, and the practical mitigations teams need now. The post Updating the taxonomy of failure modes in agentic AI systems: What a year of red teaming taught us  appeared first on Microsoft Security Blog.