Opus 5 may have solved browser-based prompt injection, the biggest security flaw haunting AI agents
AI Summary
Opus 5, when combined with Auto Mode, achieved a zero percent prompt injection success rate for browser agents in 129 tests. Without these protective measures, the success rate was notable at 3.7 percent, suggesting Anthropic might have addressed a major security issue in AI browser agents.
From the source
Opus 5 combined with Auto Mode hits a zero percent prompt injection success rate for browser agents across 129 test scenarios. Without those extra protection layers, the rate is 3.7 percent. If these numbers hold up in practice, Anthropic may have cracked one of the biggest security problems facing AI agents that operate in browsers. The article Opus 5 may have solved browser-based prompt injection, the biggest security flaw haunting AI agents appeared first on The Decoder.
The full text couldn't be loaded here (the source may require a subscription).
View original at The Decoder