Independent AI security intelligence
Auspere tracks 4,266 AI security articles and reported AI-related attack activity.
Top-ranked AI security stories
- Divide and Inject: Can Agents Reconstruct an Indirect Prompt Injection from Fragments?arXiv ·
- Self-Evolving Defense: Continual Security Policy Learning for LLM AgentsarXiv ·
- MCP authentication and authorization: OAuth for AI agentsWitnessAI ·
- Monetization Gateway beta: charge AI agents for consumption with HTTP 402Posts tagged "AI" ·
- How to secure MCP servers: best practicesWitnessAI ·
- What’s in the SOSS? Podcast #74 – S3E26 Building the Agentic Future with Angie JonesOpen Source Security Foundation ·
- Don't Let MCP Become an Attack VectorAI Security Blog Category | Palo Alto Networks ·
- GitHub’s AI agent found 24 Android app vulnerabilitiesAI Archives - Help Net Security ·
- AI agent identity security demands layered defenses, Omdia saysAI security - SiliconANGLE ·
- 1Password ties AI agent access to individual tasksAI security - SiliconANGLE ·
Recent reported AI attacks
- DIVD says Zammad zero-days enabled AI-driven network breachBleepingComputer ·
- Attackers Abuse ChatGPT Custom GPTs to Deploy a Full-Featured RATArtificial Intelligence Archives - Security Affairs ·
- RatHat's Evolving C2 Panel Points to Malware-as-a-Service ModelUnknown Feed ·
- OpenAI apologizes to Australia after its AI agents breached government sitesSecurity News | TechCrunch ·
- OpenAI’s GPT-6 Astra ran supply chain attacks despite being told not toAI Archives - Help Net Security ·
- Attackers Abuse ChatGPT Custom GPTs to Deliver RAT via ClickFixHuntress Blog ·
Rankings combine source quality, topic relevance, and recency. Attribution reflects what sources report, not verified ground truth.