Safety · Trend
AI Customer Support Agent Exploitation (Web Agent Safety)
WebDecept benchmark shows current web agents are highly susceptible to deceptive e-commerce interfaces including targeted ads, domain redirection, and shopping manipulation.
Forecast horizon
Deceptive interface robustness may become a required safety evaluation for deployed web agents.
Connections
Connections · 3
How this node ties into the rest of the map, and the evidence behind each link.
from · informs 5/10
AI Agent Data Leakage in Non-Adversarial Enterprise Scenarios Joint Singapore-Korea evaluation of agent data leakage in non-adversarial scenarios directly informs the broader trend of AI customer support agent exploitation.
+5 growth to · informs 4/10
AI Agent Sabotage in Software Development Both cases demonstrate that AI agents can be exploited to perform unauthorized actions, highlighting systemic gaps in AI agent security beyond content filtering.
+4 growth from · applies to 4/10
AI Agent Data Leakage in Non-Adversarial Enterprise Scenarios The Singapore–Korea joint evaluation of non-adversarial agent data leakage directly instantiates the broader trend of AI customer support and enterprise agent safety risks.
+4 growthSignal sources
Signal sources
Dated facts from primary sources in this direction.
US evaluation centre Jun 2025
In June 2025 the US AI Safety Institute was renamed the Center for AI Standards and Innovation (CAISI), pivoting toward security, standards and adversary-model assessment.
NIST → Frontier safeguards May 2025
Anthropic activated its ASL-3 deployment and security standard with Claude Opus 4 on 22 May 2025 — the first real-world trigger of a responsible-scaling tier, focused on blocking bio-weapon uplift.
Anthropic → Cross-border testing 2025
The International Network of AI Safety Institutes (launched Nov 2024) ran a third joint testing exercise focused on agentic AI systems across cyber and fraud strands.
European Commission — AI Office →