The Case Against AI Detectors

With the rise of large language models (LLMs) like ChatGPT, Claude, and Gemini, companies have introduced AI detection tools that claim to help educators determine when an LLM has been used to complete an assignment. Here are four reasons the Hastings Initiative does not support the use of such tools:

1. Unreliability

Performance varies across text length, genre, and hybrid texts (). AI detectors also often produce vastly different scores on repeated analysis of the same text, meaning educators cannot make consistent or defensible judgements based on their outputs ().

2. Bias Against Non-Native Speakers

A Stanford study found that AI detection tools misclassify texts written by non-native English speakers more frequently than text by native English speakers (). As a result, they can unfairly penalize students who are nonnative English speakers.

3. Atmosphere of Distrust

AI detection-focused approaches can foster an environment of distrust and anxiety in the classroom, undermining educational relationships ().

4. No Sliver Bullet

With educator adoption of AI detection tools, there is a growing student market for tools that “humanize” text generated by AI to bypass detection (). As LLMs and humanizer tools continue to improve, detection tools cannot solve academic integrity concerns.

��vlog

Hastings Initiative for AI and Humanity