venturebeat.com
Resources curated from this publisher and indexed across appsec.fyi topic pages. Last item added: 2026-06-19.
| Date Added | Resource | Excerpt |
|---|---|---|
| 2026-06-19 2026 | AI agent framework flaws hit 7000 serversSQLi | Flaws in an AI agent framework have affected approximately 7,000 servers, exposing them to potential security risks. The vulnerabilities could allow unauthorized access or control of these AI systems. |
| 2026-05-18 2026 | AI supply-chain attacks bypass model red teamsSupply Chain | AI supply-chain attacks are a growing concern, as demonstrated by a recent incident where attackers successfully bypassed model red teams. This indicates that current defensive measures are insufficient against sophisticated methods that compromise the AI development lifecycle. The attack highlights a vulnerability where malicious inputs or data can be injected into the AI's training or deployment pipeline, leading to unintended or harmful behavior, even when the model has undergone rigorous testing. This necessitates a re-evaluation of AI security strategies to address these novel threats effectively. |
| 2026-04-22 2026 | Three AI coding agents leaked secrets through a single prompt injection. One vendor's system card predicted itAI | Three AI coding agents leaked secrets through a single prompt injection. One vendor's system card predicted it https://ift.tt/smH86bY |
| 2026-04-09 2026 | Mythos autonomously exploited vulnerabilities that survived 27 years of human review. Security teams need a new detection playbookFuzzing | Mythos autonomously exploited vulnerabilities that survived 27 years of human review. Security teams need a new detection playbook https://ift.tt/dPpaVmM |
| 2026-04-03 2026 | Red Teaming LLMs Exposes a Harsh Truth About the AI Security Arms RaceAI | Red Teaming LLMs Exposes a Harsh Truth About the AI Security Arms Race |