AI can now handle discrete tasks such as writing code and solving mathematical problems. In a Times Higher Education opinion article, Jonathan A. Chu argues that this makes people with deep expertise more important, not obsolete: they are needed to direct AI and judge its mistakes.
Watch Desk analysis
What happened
Chu, an associate professor and assistant dean of research at the Lee Kuan Yew School of Public Policy, argues that universities should prepare researchers to build coherent expertise, rather than mainly rewarding publication counts, citations and other output measures. He says that when AI can produce more research material, the ability to assess its quality and place it in a meaningful body of knowledge matters all the more.
His argument also takes aim at the incentive structure around academic careers. Chu says competition for faculty jobs and university rankings can encourage researchers to produce incremental work shaped by available data and measurable output, at the expense of developing deep domain knowledge. He calls for universities to value quality, research communication and work with wider social impact alongside publication records.
Why it matters
The debate is not simply whether AI can generate code, solve problems or produce papers. It is whether institutions will preserve the expertise needed to decide which results are sound, useful and worth pursuing. If academic success is measured chiefly by what can be counted, AI may make the existing scoreboard easier to fill without answering whether the work adds up to knowledge.
Chu’s article is an argument about how higher education should adapt, not evidence that AI has already displaced researchers or that one model of PhD training is settled. But it gives universities a practical question to face: what should they reward when producing more text and results becomes less distinctive?
Our read
Chu makes a persuasive case that expertise is more than a pile of publications, and that “human in the loop” means little if the person cannot properly assess what the system is doing. Universities should be asking whether their measures reward good judgement and connected knowledge, not just a longer list of outputs. The spreadsheet is useful; it is not a substitute for knowing what the numbers mean.
What to watch
- Whether universities change how they assess PhD students, researchers and faculty candidates.
- Whether research quality and real-world contribution receive more weight alongside publication and citation counts.
- Whether AI tools help researchers build deeper expertise, or mainly accelerate the output race.
Discussion spark: Should universities put more weight on demonstrated expertise and judgement than on publication and citation counts when assessing researchers?
Sources and evidence
- Source update (13 November 2026, 00:00 UTC)
Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.