Retract or be damned: a dangerous moment for science and the public

Dear Editor

We firmly agree with the Editorial "Retract or be damned: a dangerous moment for science and the public". What concerns us is that this moment is growing more dangerous because of a bystander effect. With scientific research visibly in difficulty, the major players who could help are largely walking on by.

Publishers expect the institutions to act to stop the poor quality and fraudulent research being created. Institutions argue that the problems should be found by better peer review, whilst they suppress scandals of fraud and misconduct by their own staff to protect their reputation. Funders aim to fund new research, not quality control. Whilst each of these major players waits for the other to act, the state of scientific research worsens, with the vital sign of publication numbers hitting a seriously unhealthy number that has not yet plateaued.

A telling example of the lack of serious action to tackle research integrity is the Dutch microbiologist Elisabeth Bik, who is an expert at catching manipulated images in scientific papers. Bik’s work is a tremendous public service, but she isn’t paid by a university or a scientific publisher, instead she is crowd funded.

Who can stop this bystander effect? Ideally a player that has huge international gravitas will step forward, which would then trigger others to act.

But this needs to be serious action. It can't be yet another special issue on research integrity and then back to business as usual. It can't be a one-day national meeting on research integrity and a report that few people may read.

Serious long-term funding needs to be committed to research quality control. We suggest that 1% of publishers profits and 1% of all scientific funding should be diverted to quality control. This 1% of investment would have massive benefits for the remaining 99%.

There is a growing field of meta-research with many talented researchers and good ideas to improve research quality (for example, the Association for Interdisciplinary Meta-Research and Open Science). But we cannot expect this small field to revive scientific research alone.

Ultimately, the power to correct the published record rests with journals and publishers. In our experience, when provided with damaging descriptions of flawed science that could arise from paper mills, some journals and publishers choose to look away.

While we have long argued that publishers must devote more resources to post-publication correction, there are currently few incentives to speed or scale retractions. There are also few consequences for journals that allow unreliable research to remain in the literature. We urgently need systems that meaningfully reward responsible journals, and decisively penalize journals that fail to act.

Failing to issue justified retractions has serious consequences. This forces the research community to build on weakened evidence, with real impacts on scientific productivity and careers. By not acting on unreliable science, journals and publishers are actively rewarding paper mills. These unintended “rewards” need to be urgently reconsidered, now that paper mills are armed with AI capacity to further scale production of fraudulent science.

We ask all institutions, publishers and funders to urgently consider how they can recalibrate the current imbalance between publication and correction, and research quantity and quality.

For the sake of science, scientists and the broader public, the time to act is now.