AI Security

Twitter

Facebook

Copy Link

Key Facts

AffectsLLMs, Autonomous Agents, Enterprise AI

Key DefenseRed Teaming, Guardrails, Monitoring

Key ThreatsPrompt Injection, Data Poisoning, Jailbreaks

Top News

Sriram Krishnan steps down as White House AI advisor, plans new policy institution

Sriram Krishnan is leaving his role as senior White House AI advisor at the end of June after helping shape the Trump administration's AI Action Plan. The former Andreessen Horowitz partner, who worked closely with AI czar David Sacks, plans to start an outside institution focused on AI policy while continuing to advise the administration on America's path to AI dominance.

Policy11 sources

Sriram Krishnan steps down as White House AI advisor, plans new policy institution

Policy11 sources

AI agent discovers 21 zero-days in FFmpeg for $1,000 as Chrome patches record 429 security bugs

A security startup's autonomous AI agent found 21 previously unknown vulnerabilities in FFmpeg for roughly $1,000 in compute costs, some hiding for over 20 years. The same week, Google shipped Chrome 149 with patches for 429 security bugs, the most ever in a single release. The developments highlight how AI is accelerating vulnerability discovery faster than human teams can address them.

Technology2 sources

AI agent discovers 21 zero-days in FFmpeg for $1,000 as Chrome patches record 429 security bugs

Technology2 sources

Self-replicating Miasma worm compromises 73 Microsoft GitHub repositories in supply chain attack

Microsoft removed 73 GitHub repositories after they were infected with the Miasma worm, a self-replicating credential-stealing malware that activates when developers open projects in AI coding tools. The attack exploited the same compromised credentials from a May breach, raising questions about incomplete credential rotation and highlighting vulnerabilities in the modern software supply chain.

Technology9 sources

Self-replicating Miasma worm compromises 73 Microsoft GitHub repositories in supply chain attack

Technology9 sources

Trump's AI strategy prioritizes military adoption and cybersecurity over industry regulation

Donald Trump signed an executive order on AI security and a national security memorandum accelerating military AI adoption. The orders establish voluntary 30-day reviews for frontier models before public release and bar AI vendors from disabling deployed military systems without approval. While acknowledging risks from autonomous weapons systems, the framework keeps light-touch regulation for commercial AI, raising questions about whether voluntary measures adequately address AI safety concerns.

Policy7 sources

Trump's AI strategy prioritizes military adoption and cybersecurity over industry regulation

Policy7 sources

AI discovers critical Zcash vulnerability, exposing privacy coin tradeoffs as ZEC crashes 40%

Anthropic's Claude Opus 4.8 helped uncover a four-year-old flaw in Zcash's Orchard shielded pool that could have enabled unlimited counterfeit ZEC creation. The disclosure triggered a 40% price crash and emergency network upgrades, while highlighting the tension between privacy and auditability in blockchain systems.

Technology3 sources

AI discovers critical Zcash vulnerability, exposing privacy coin tradeoffs as ZEC crashes 40%

Technology3 sources

Anthropic and Trump Administration Ease Tensions Despite Pentagon's Ongoing Blacklist Battle

The months-long dispute between AI company Anthropic and the Trump administration is showing signs of thawing, even as the Pentagon maintains its supply-chain risk designation. Anthropic CEO Dario Amodei has engaged with White House officials, while the NSA reportedly deploys the company's advanced Mythos AI model for offensive cyber operations.

Policy4 sources

Anthropic and Trump Administration Ease Tensions Despite Pentagon's Ongoing Blacklist Battle

Policy4 sources

NSA deploys Anthropic's Mythos AI model for cyber operations despite Pentagon blacklist

The NSA is using Anthropic's Mythos AI model for offensive cyber operations, with half a dozen Anthropic engineers embedded inside the agency. This development comes despite the Department of Defense designating Anthropic a supply chain risk and banning its technology after the company refused to allow use of its AI for mass surveillance and autonomous weapons.

Technology4 sources

NSA deploys Anthropic's Mythos AI model for cyber operations despite Pentagon blacklist

Technology4 sources

Synology transforms DiskStation Manager into intelligent data platform with private AI workflows

Synology showcased its next-generation DiskStation Manager at Computex 2026, evolving from a storage operating system into an intelligent data platform. The update emphasizes on-premises AI workflows and local AI agents that process data without cloud dependencies, addressing enterprise concerns around data control, privacy, and compliance while enabling automated administrative tasks and AI-driven insights.

Technology3 sources

Synology transforms DiskStation Manager into intelligent data platform with private AI workflows

Technology3 sources

Prev 1…9 10 1112Next