Safety · Trend
AI-Enabled Persuasion Capabilities
AISI large-scale experimental study of conversational AI persuasion levers, published in Science.
Forecast horizon
Persuasion evals will become standard national-security and platform-risk metrics.
Connections
Connections · 8
How this node ties into the rest of the map, and the evidence behind each link.
from · applies to 6/10
Persuasion Index (PI): Theory-Guided Rhetorical Analysis Framework The Persuasion Index provides a theory-guided framework for auditing AI-mediated persuasion, directly relevant to AI persuasion capability assessment.
+6 growth from · informs 4/10
LLM Psychological Manipulation in Multi-Turn Interactions CogManip benchmark findings on covert manipulation strategies in multi-turn LLM interactions directly inform research on AI-enabled persuasion.
+4 growth from · drives 4/10
Philosophical Vertigo from Human-AI Interaction Emotionally salient persuasive AI conversations are a pathway into ontological shock and epistemic destabilisation.
+4 growth from · applies to 4/10
Philosophical Vertigo from Human-AI Interaction Persuasive long emotionally salient AI dialogues are a pathway into ontological shock and epistemic destabilisation.
+4 growth from · requires 4/10
Philosophical Vertigo from Human-AI Interaction Persuasive long-form AI dialogue is a pathway into ontological shock and epistemic destabilisation.
+4 growth from · tracked by 3/10
UK AI Safety Institute AISI conducted and published large-scale experiments on AI persuasion capabilities in Science.
+3 growth to · enables 3/10
Persuasion Index (PI): Theory-Guided Rhetorical Analysis Framework The Persuasion Index provides a theory-grounded measurement tool for auditing AI persuasion capabilities identified as a safety concern.
+3 growth from · frames 3/10
Industry-Wide Jailbreak Severity Scoring Framework Jailbreak severity scoring is a governance response to AI persuasion and manipulation capabilities.
+3 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 →