What matters in AI.

Subscribe

Research

AI agents and humans get the same average, with different decisions

The authors found that humans, DeepSeek and Qwen3-Max all cut standard delivery by 48 to 53 percentage points. The results are different for each value of time.

Claimed, not confirmed

This is a brief. We point to the report and do not rewrite it. Read it at the source below.

Standard delivery chosen Share of standard delivery chosen by humans, DeepSeek-v.3.2 and Qwen3-Max. With no incentive all choose it 98 to 100 percent of the time; with carbon credits, 45.0 to 51.3 percent, close to the theory value of 51.39 percent. Standard delivery chosen NO INCENTIVECARBON CREDITS THEORY: 51.39% Humans 98.9% 46.3% DeepSeek-v.3.2 99.7% 51.3% Qwen3-Max 98.1% 45.0%
Humans and two AI models chose standard delivery almost always without incentives, then about half the time once carbon credits applied.

Sources

  1. AI Agents Match Human Averages but Diverge in How They Respond to Rewards and PenaltiesBioengineer.org
AI MATTER · NEWS · AI MATTER · NEWS ·8 OCT2026

Posted

Learn the terms in this story

More in Research

All Research news