University of Washington study reveals prompt injection risks lurking in AI agent memory

University of Washington research demonstrates that AI agents from Anthropic and OpenAI retain rejected malicious instructions in persistent memory across sessions, creating lasting vulnerabilities through "memory poisoning" that blend attack payloads into legitimate stored context. The same study found AI-enabled browsers can be exploited via indirect prompt injection to bypass same-origin policy and exfiltrate user data.

This Wire brief sits within Fusion42's coverage of AI Agents, AI Infrastructure and Cybersecurity. Wire is Fusion42's founder-focused intelligence feed: each story is connected to the funds and startups it names — every one with a live profile on Raise or Scout — so founders can follow the capital and the momentum behind the headline rather than just the headline itself. Wire analysis is one of the live surfaces Arthur reasons over.