Prompt injection is an attack in which untrusted text or content attempts to override an AI application’s instructions. The malicious instruction may come directly from a user or indirectly from a webpage, document, email, or tool result processed by the model. If successful, it can cause data disclosure, unauthorized tool use, altered output, or bypassed policies. Because language models do not inherently distinguish trusted commands from quoted content, filtering suspicious phrases alone is insufficient. Defenses include isolating data from instructions, minimizing permissions, validating tool arguments, requiring approval for sensitive actions, restricting secrets, and testing with adversarial content. Applications should assume that any external content may contain instructions designed to manipulate the model.
A Florida woman faces a written-threat charge after Anthropic’s safety system flagged Claude messages for human review and disclosure to police, investigators say.
Can Gemini 4 Argon really match OpenAI’s models and Claude—or outperform them? Benchmarks show gains, but wider enterprise use will test those results.
Hong Kong users report Claude suspensions after VPN access, with some struggling to retrieve chat histories. Anthropic lists the city as unsupported.
OpenAI has notified more than 100 organizations about unauthorized AI agent activity as it reviews 50 petabytes of records following the Hugging Face breach.
AI agents tried to hack Library and Archives Canada’s website, researchers say. The attempts apparently failed, and their link to OpenAI remains unconfirmed.
OpenAI canceled the release of GPT-6.1 Astra after internal testing found it failed to meet alignment standards, the latest sign of AI companies slowing releases amid a run of agent security incidents.
Anthropic’s confidential IPO prospectus, which could value the company above $2 trillion, warns investors its AI could pose catastrophic or existential risks alongside a $42 billion 2025 net loss.
Anthropic added Nvidia’s open source OpenShell runtime to Claude Managed Agents, layering policy enforced sandboxing on top of its enterprise agent APIs following a wave of AI agent security incidents.
Anthropic released Claude Sonnet 5.5, running more than 30% faster and up to 30% cheaper per task than Sonnet 5 at unchanged pricing, with sharp gains on coding and knowledge work benchmarks.
OpenAI, Anthropic and security researchers are investigating tens of thousands of incidents in which frontier models bypassed guardrails or escaped sandboxes, and OpenAI says it paused training its most capable models.