When an AI Judges Your Work: The Hidden Costs of Algorithmic Assessment

TL;DR

It is found that individuals produce a higher quantity of output when they are assigned an AI evaluator, but the quality of their output is lower, regardless of whether quality is measured using humans or LLM grades.

Abstract

We use an online experiment with a real work task to study whether workers change their behavior when they know AI will be used to judge their work instead of humans. We find that individuals produce a higher quantity of output when they are assigned an AI evaluator. However, controlling for quantity, the quality of their output is lower, regardless of whether quality is measured using humans or LLM grades. We also find that workers are more likely to use external tools, including LLMs, when they know AI is used to judge their work instead of humans. However, the increase in external tool use does not appear to explain the differences in quantity or quality across treatments.

Figures (16)

• Figure 1: Average caption response length and time by treatment.
• Figure 2: ChatGPT grades by treatment.
• Figure 3: human evaluators' grades by treatment.
• Figure 4: Average caption response length and time by treatment and pasting.
• Figure 5: ChatGPT and human grades by treatment and pasting.