A.I. Has a Measurement Problem

There’s a problem with leading artificial intelligence tools like ChatGPT, Gemini and Claude: We don’t really know how smart they are.

That’s because, unlike companies that make cars or drugs or baby formula, A.I. companies aren’t required to submit their products for testing before releasing them to the public. There’s no Good Housekeeping seal for A.I. chatbots, and few independent groups are putting these tools through their paces in a rigorous way.

Instead, we’re left to rely on the claims of A.I. companies, which often use vague, fuzzy phrases like “improved capabilities” to describe how their models differ from one version to the next. And while there are some standard tests given to A.I. models to assess how good they are at, say, math or logical reasoning, many experts have doubts about how reliable those tests really are.

This might sound like a petty gripe. But I’ve become convinced that a lack of good measurement and evaluation for A.I. systems is a major problem.

For starters, without reliable information about A.I. products, how are people supposed to know what to do with them?

I can’t count the number of times I’ve been asked in the past year, by a friend or a colleague, which A.I. tool they should use for a certain task. Does ChatGPT or Gemini write better Python code? Is DALL-E 3 or Midjourney better at generating realistic images of people?