of Economics (QJE), observes a change in the papers which are being submitted: “the figure format, some of the analysis, the more polished text”. In July about one in four working papers filed with the National Bureau of Economic Research, an American repository, contained AI-assisted writing, according to an analysis by Paul Goldsmith-Pinkham of the Yale School of Management. As papers become easier to write, some journals fear being inundated. The Journal of Political Economy (JPE), another top-tier outlet, will increase its submission fee to as much as $500 per paper in 2027 to discourage slop. It is not just the prose. The analytical core of economics papers can increasingly be done by AI agents. David Yanagizawa-Drott of the University of Zurich has started a research programme dubbed Autonomous Policy Evaluation (APE). Like human economists before it, Dr Yanagizawa- Drott’s AI uses publicly available data to gauge the effect of policies. Among other things, it found that Switzerland’s ban on second homes substantially lowered investment in residential and commercial construction, and that Egypt’s large currency depreciation in 2016 affected consumer- good producers more than capital-goods makers. Is agentic research any good? Dr Yanagizawa-Drott and his colleague, Olaf Willner, used an AI verifier to check 1,000 APE papers. None was error- free; 842 contained at least one severe error, including ones invalidating headline results. To see roughly how Ape stacks up against human primates, The Economist teamed up with Refine, the research-verification firm founded by Dr Golub. Refine chose 100 papers at random from those accepted for publication in the top five journals (including QJE and JPE) in the first nine months of 2026. It then subjected them to thorough review using several frontier AI models plus a once-over for serious errors by a human reviewer. The exercise flagged 4,158 issues in the 100 papers. One in ten papers contained at least one grave error (other problems concerned interpretation or inconsistencies between tables and text). The two exercises are not directly comparable, but human economists’ lower error rate might give them some hope. In ideas, too, people retain the edge. One experiment, by Alexis Akira Toda of Emory University, found that when considering papers on economic theory alone, AI lowers the cost of checking for a candidate flaw but “does not yet reliably substitute for the

human judgment that locates the flaw in the first place”. Another, which compared APE papers with published papers found the difference is mostly in ideas, not execution. In a further study, Michele Zampa of the Geneva Graduate Institute used an AI referee to grade 981 AI papers and 41 human ones on things like the novelty of research questions and acknowledgment of methodological limitations. AI did better in crowded fields with lots of existing research to digest. Humans performed at the same level regardless of the size of the literature. A related worry is that AI produces plausible-looking results by brute force, slicing and dicing data in hundreds of ways until something statistically significant pops up. Such “p-hacking” can be contained by checking how sensitive results are to changes in data or methods. Nic Fishman of Harvard and Gabriel Sekeres of Cornell University calculate that the number of needed robustness checks would rise from 50 for a human author to an impractical 7,000 for agentic ones. No wonder the APE project has a warning label not to use its papers for evidence-based policymaking. The machines are gaining ground, however. Benjamin Kohler of ETH Zürich, another Swiss university, and colleagues, let AI reproduce 48 social- science papers, using only the studies’ raw data and a description of the method. The discrepancies between the original papers and the AI versions were due in roughly equal measure to problems with the AI and human work. But the better the AI model, the more the balance tilted against humans. As it gets less error-prone and more creative, AI will remove the drudgery from parsing historical data, make complex modelling as easy as vibe- coding, identify cherry-picked numbers and improve economics in countless other ways. Already it is more than a tool. “We need to think about AI as a co-author,” says Dani Rodrik of Harvard, though it will be a while before economics can dispense with human co-authors altogether. “There is still a large amount of uncertainty about the right models, how to extrapolate from the evidence and what the right research frame is.”

Keynes would agree. A master-economist, he noted, “must be mathematician, historian, statesman, philosopher…He must understand symbols and speak in words. He must contemplate the particular in terms of the general, and touch abstract and concrete in the same flight of thought.” It may be a while before AI meets Keynesian criteria. Then again, as Keynes also noted, not many humans meet them, either. ■ This article was downloaded by zlibrary from https://www.economist.com/finance-and-economics/2026/11/19/ai-is-swallowing-maths- is-economics-next

· Science & technology

America has slashed funding for biomedical science abroad Some grasshoppers fly like butterflies to mislead predators SpaceX achieves another historic first LLMs are developing filthy mouths Should women sync their lifestyle to their menstrual cycle?

Science & technology | Research and destroy America has slashed funding for biomedical science abroad It has also weakened oversight of how future grants are spent Nov 19th 2026 PAOLO DECUZZI has spent decades trying to get tiny devices to do big things. The biomedical engineer, based at the Italian Institute of Technology in Genoa, has devised an assortment of nanotechnologies that aim to deliver drugs more effectively. To turn these inventions into working medical treatments, he has long collaborated with American scientists on projects funded by the National Institutes of Health (NIH), the largest funder of biomedical research in the world. But last year the NIH abruptly suspended research to study a microscopic plate Dr Decuzzi had invented to deliver drugs directly into arthritic joints. He estimates that another one, intended for cancer patients, has been set back by as much as two years.

Many researchers have experienced something similar. In the fiscal year ending September 2024 the NIH gave an estimated $400m to 3,600 collaborating foreign scientists. It was a fraction of the nearly $35bn in grants the agency doled out, but an easy target for the Trump administration as it cut NIH projects broadly. The result is that America’s system for funding international research has been gutted. Extensive internal NIH data shared with The Economist on the condition of anonymity shows the scale of the change. In the fiscal years 2022-2024 the agency awarded an average of 2,040 grants a year to projects with foreign components. The average for 2025 and 2026 was 38% lower (see chart 1). In its purported pursuit of transparency and accountability, the administration has disrupted valuable science and inadvertently made its spending more inscrutable. For all of America’s science dominance, some resources cannot easily be found at home. Dr Decuzzi’s microparticles, for example, cannot be bought off the shelf. Conducting clinical trials on rare diseases often requires enrolling patients from countries where cases are more common. Such collaborations have long been governed by strict conditions, at least on paper. The existing guidelines were blown apart in May 2025, however,

when the NIH barred American universities from using an established scheme to pass grant money to their overseas partners. Projects that already had funding could continue, but those waiting for their next yearly instalment were suspended until the following spring. For over a year, the NIH essentially shut off its means of funding new international research partnerships. The architect of this overhaul is Jay Bhattacharya, who has led the NIH since April 2025. He is also a vocal proponent of the covid lab-leak hypothesis, which contends that the virus escaped from the Wuhan Institute of Virology in China. (America’s National Intelligence Council has concluded, albeit with low confidence, that a natural origin was more likely.) The lab had previously received NIH funding, which troubles Dr Bhattacharya. “The goal”, he said in a 2025 interview with Christianity Today, an Evangelical magazine, is “to put those foreign collaborations on more solid footing so that a Wuhan situation never happens again”. Concerns about oversight are not unfounded. After the pandemic the institute in Wuhan refused to co-operate with its American research partner, leaving investigators unable to retrieve records of NIH-funded research that could help establish the virus’s origins. In 2023 the Government Accountability Office (GAO), a congressional watchdog, found that the NIH “had not taken near-term steps to manage risks” associated with funding research abroad, and made recommendations that the agency then acted on. The Trump administration has gone much further. Although foreign grants make up a small share of NIH awards, they have suffered disproportionately. At the National Institute on Ageing grants with a foreign component fell by 49%, compared with 11% for domestic grants. At both the children’s health and biomedical imaging institutes foreign grants declined by over 40% even as domestic awards rose slightly (see chart 2). All told, grants of $5m or more that include a foreign component fell from an average of 20 in 2022- 2024 to just three in 2026.

The reductions cut across nearly every institute but those studying infectious diseases and mental health saw the largest drop in the number of projects. The Economist’s analysis reveals that the number of grants whose title mentions Africa, clinical trials, children, mothers, HIV or “ageing and dementia” was roughly half that of the average during the previous three years. For Dr Decuzzi the chaos means years of lost postdoctoral students, slower discoveries and more paperwork. “Once we have destroyed things, rebuilding is actually very difficult,” he laments. For others the consequences have been grimmer still. Dvora Joseph Davey, an epidemiologist studying HIV at the University of California in Los Angeles, has long been affiliated with the University of Cape Town. In May 2025 the NIH pulled her international funding, forcing her to end a clinical trial. She had been investigating how best to prevent HIV transmission among pregnant and breastfeeding women in a country where roughly 20% of women of reproductive age have the virus. Study participants subsequently lost access to PrEP, drugs that reduce the risk of HIV infection by 99%. Two women in her study who stopped taking the drug became infected and one of them transmitted the virus to her baby. Dr Bhattacharya calls this case “unfortunate”, but says that ensuring patient care following a grant suspension is the researchers’ responsibility.