Skip to content

Briefings · Judgement

Three mathematicians on Terence Tao's blog argue AI has exposed a flaw in how experts value their own work

A guest essay on Terence Tao's blog argues that measuring experts by output made them easy to compare with machines, and AI has only exposed it.

by ·3 min read·

New here? Start with the free AI Survival Kit →

Three mathematicians on Terence Tao's blog argue AI has exposed a flaw in how experts value their own work

0:00 / 5:35

This voice is generated by AI.

Part of the guide Will AI take your job? How to find out for your role, in an afternoon

Three mathematicians on Terence Tao's blog argue AI has exposed a flaw in how experts value their own work

Terence Tao's blog, What's new, published a guest essay on 2 November by Karthik C. S., Amit Sahai and Jie Xue titled "Egoless Mathematics in the AI Era". The authors trace mathematicians' anxiety about AI to a habit that started before the machines arrived. "We increasingly measure intellectual life through publications, citations, awards, and other indicators of performance," they write. Once those measures start to define what the work is worth, people come to look like "collections of productive abilities", and any machine that performs the tasks better "appears to threaten our worth". Their verdict: "the problem began before AI arrived. We had already come to understand ourselves in terms suited to machines."

The refuge the authors close off

The usual advice is to find what is distinctively human, such as creativity, intuition or taste, and build your position there. The authors reject it: "If we value these abilities primarily because they give us an advantage over machines, we remain trapped in the same conception of intellectual worth."

That sits close to The War Room's own position, which is that the dividing line runs through judgement and accountability. The essay makes the point sharper. If you hold judgement only as a competitive edge, it becomes one more output metric. Output metrics are exactly what machines are built to beat.

The authors' alternative rests on the gap between a correct derivation and its significance. "An axiomatic system permits indefinitely many deductions, but the ability to produce them does not by itself supply a worthy direction of inquiry." Choosing which definition will clarify things, and which theorem reveals something fundamental, is a matter of direction. They also note that being able to carry out longer chains of reasoning does not by itself explain understanding.

The same holds outside mathematics. In any professional field, a system can now produce correct work in volume. Deciding which work matters, and answering for that decision, is a separate job. It is also the job that keeps its value.

The authors used the machines

The essay is open about its own use of AI. Its statement of AI usage thanks GPT Astra 6.0 and GPT 6.1 Sol for help with drafting, and says "the ideas expressed here are entirely our own". Tao adds that the post was written in a different file format and converted using AI. The authors also decline to take sides between mathematicians who want to use AI under agreed principles and those who prefer not to use it at all. In their view, both positions can reflect the same commitment to mathematical truth.

That is the working pattern. AI does the routine work. You do the thinking. The drafting and the file conversion went to the machine. The argument stayed with the three people who signed it.

For an operator, the lesson is to look at where your sense of value comes from. If it comes from volume, any tool that produces more volume will look like a rival. If it comes from choosing direction and owning the result, the same tool makes you better at what you already do.

Your Next Move

  1. Audit your scoreboard. Write down the three numbers you would cite in your next review. If they count output, such as reports filed, tickets closed or decks shipped, they set you up for a direct comparison with a system that produces more. Add one direction decision for each quarter: a problem you chose to take on, or one you dropped, and your reasons. To sort which parts of your week a system can absorb, use our method for scoring your role task by task.

  2. Hand over the derivations. Choose one recurring task where the result is easy to check and the burden is volume. Give the drafting to an AI tool. Keep the review, the framing and the sign-off for yourself.

  3. Write a usage statement. Copy the authors. Add one line to your next major deliverable saying what the tool did and what you decided. That makes your judgement visible and puts your name on it, which is where it belongs.

I build systems like the one publishing this site. → Work with me

About the author

Jo

Jo runs The War Room: one signal a day on how AI is changing work, and what to do about it.

Sources

Get the Briefing

Want more intelligence like this?

The weekly briefing, free — and the AI Survival Kit with it.

Takes 30 seconds. No spam.

More in Briefings

See all