Can you spot the expensive AI?
The Blind Test challenges users to distinguish between AI-generated and human-created content across images, text, and audio. Participants are tested on their ability to spot which works were produced by expensive AI models versus those made by humans.
Background
- This is a web-based "blind test" game that challenges you to distinguish between responses generated by a cheap AI model, an expensive AI model, and a human.
- The test reflects a growing real-world concern: as AI language models become cheaper and more capable, the line between human-written and machine-generated text keeps blurring.
- "Expensive AI" typically refers to large, state-of-the-art models like GPT-4 or Claude 3 Opus, which cost more per query; "cheap AI" refers to smaller or older models like GPT-3.5 or Llama 2, which are much cheaper to run.
- The game is a playful but pointed way to ask whether the premium you pay for top-tier AI actually buys text that is noticeably better or more "human" — or whether the difference is vanishing.