Cybersecurity · 44 views
Prompt injection breaks today’s AI agents, study warns
Today’s AI web agents have no dependable defenses against prompt injection, according to new research showing that not a single attack scenario was consistently blocked across leading systems powered by GPT‑5 and Gemini.
Summary
Today’s AI web agents have no dependable defenses against prompt injection, according to new research showing that not a single attack scenario was consistently blocked across leading systems powered by GPT‑5 and Gemini. The findings come from StakeBench, a stakeholder-centric benchmark developed by researchers from Nanyang Technological University, ST Engineering, IBM Research, and the University of Illinois Urbana-Champaign to evaluate prompt injection attacks against AI agents operating in realistic web environments. The researchers executed 3,168 adversarial runs across NanoBrowser and BrowserUse using 264 benchmark cases.
AI summaries can be wrong sometimes—always verify important details using the source article.
How AI & Automation are usedMore from Cybersecurity
Continue reading recent Cybersecurity coverage
- 10 most powerful cybersecurity companies todayContinue reading
- Bright Security Introduces STAR Fuzzer, Agentic Fuzzing That Writes Its Own Harnesses and Proves Every FindingContinue reading
- Malicious File in Downloads Folder Tries to Trick AI Coding Agent Into Installing MalwareContinue reading
- Unpatched Tailscale SSH Flaws Could Let a Hacked Device Reach Root on Other ServersContinue reading
Support HappeningNow
Independent AI-powered news analysis is reader-supported. Your contribution helps cover infrastructure, summaries, and continued platform development.
Support HappeningNow