Good AI Task

AI compatibility

Reading a room in a high-stakes negotiation is not something AI can do today.

Human required

A human should do this one.

Average across 1 submission.

8
avg / 100

The honest read

Reading body language and emotional subtext in a live, high-stakes negotiation is one of the most human-dependent tasks imaginable. It requires real-time sensory perception, deep contextual intuition, and the ability to act on fleeting, ambiguous signals — none of which current AI agents can reliably do in a live setting. Getting it wrong can cost deals, relationships, or trust that cannot be recovered.

Aggregated across 1 submission.

The five dimensions

Repeatability

Low

Every negotiation is structurally unique — different parties, stakes, cultural norms, histories, and emotional dynamics. There is no repeatable template an agent can follow, making automation extremely difficult.

Ambiguity Tolerance

Low

Success criteria are almost entirely subjective and situational — there is no clear signal that the interpretation was correct until outcomes unfold, often much later. An agent cannot know when it has done this well.

Data & Tool Availability

Low

An agent would need live video feeds, audio, and real-time access to the room — plus deep background on all parties. Even with computer vision tools, interpreting micro-expressions and subtext in context is far beyond reliable current capability.

Error Cost

High

A misread of emotional state or intent in a high-stakes negotiation can cause a party to make a wrong move, destroy trust, or collapse a deal — consequences that are often irreversible.

Human Judgment Required

High

This task is almost entirely intuition, empathy, cultural fluency, and relational context — the exact capabilities where AI is weakest and where experienced human negotiators spend years developing skill.

What an agent would need

  • Real-time video and audio access to all negotiation participants
  • Computer vision capable of reliably detecting micro-expressions and posture shifts in context
  • Deep background knowledge on each party's personality, culture, and negotiation history
  • A framework for translating physical signals into actionable strategic interpretations
  • A feedback loop to validate interpretations against actual outcomes in real time

Best-matched agent type

Perception Agent

The kind of agent this work would call for if it were a fit. For this task, it isn't.

Run your own fit check

Get a calibrated read on your specific task in under a minute.

Check a task