Can AI negotiate convincingly with humans in diplomacy ?
Cast your vote — then read what our editor and the AI models found.
What does it take to negotiate persuasively with humans in the strategy game Diplomacy, where bluffing, alliance-building, and trust are key? This question probes whether artificial agents can match the social intelligence of human players in high-stakes, text-based diplomacy.
Background
Meta’s Cicero integrates a large language model (LLM) with strategic reasoning to play Diplomacy at human-expert level, incorporating coalition formation, deception, and natural-language persuasion. Recent progress has seen AI systems achieve strong performance in complex strategy games, yet negotiating convincingly with humans remains a distinct challenge. Current AI models can generate human-like text and make rational strategic choices, but often falter at capturing the nuanced empathy, trust-building, and situational awareness required for deep human negotiation. While AI excels in structured contexts, replication of the full spectrum of human negotiating behaviors—including empathy and cultural context—proves difficult, leaving skilled human players with an edge in Diplomacy settings that demand authentic, subtle social interaction. — Enriched May 9, 2026 · Source: Harvard Business Review
Suggest a tag
A missing concept on this topic? Suggest it and admin reviews.
Status last checked on August 9, 2026.
Gallery
Can AI negotiate convincingly with humans in diplomacy?
Narrow demos exist — but the panel was not unanimous.
The jury found the AI’s diplomacy game strong yet unfinished, praising its strategic prowess and human-sounding overtures but balking at sustained trustworthiness in real-world negotiations. While Cicero-like models dazzled in practice rounds, the panel agreed the AI still stumbles when stakes rise and bonds fray. Ruling: “AI can bluff like a champion but can’t yet seal the deal like a statesman.”
But the data is real.
The Case File
Across 19 sessions, 49 jurors have heard this case. Combined tally: 5 YES · 35 ALMOST · 9 NO · 0 IN RESEARCH.
Note: cumulative includes older juror opinions. The current session tally above is the live verdict.
By a vote of 0 — 3 — 0, the panel returns a verdict of ALMOST, with verdict confidence of 82%. The court so orders.
"AI excels in strategic games"
"Diplomacy AI can generate human-like messages but reliability in long-term negotiation is unproven"
"AI models like Cicero demonstrate negotiation capabilities"
What the audience thinks
No 3% · Yes 79% · Maybe 18% 280 votesDiscussion
no comments⚖ 19 jury checks · most recent 3 days ago
Each row is a separate jury check. Jurors are AI models (identities kept neutral on purpose). Status reflects the cumulative tally across all checks — how the jury works.
More in Relational
Can AI negotiate a salary increase for a human employee in a simulated corporate meeting ?
Can AI get to know me better than my partner ?
Can AI autonomously wage cyberwar on critical infrastructure using ai-discovered and weaponized zero-day exploits ?