korrents
Derek Lowe

What Derek Lowe thinks about AI and science

@derek-lowe · 16 positions · 0 changes of mind

Medicinal chemist. He has worked on preclinical drug discovery in the pharmaceutical industry since 1989, across several companies and therapeutic areas, and has written the drug-discovery blog 'In the Pipeline' since 2002 - the longest-running blog in the field, editorially independent and hosted by Science.

Derek Lowe did not write this page.

We collected these quotes from things they published elsewhere, and every quote links to where it was said. They have no account here and have not endorsed this site. Quotes are word for word; the short line under each one is our own restatement, not their wording. Their own site. Is this you? Claim it or ask us to remove it. Or tell us what is wrong here.

6 dated positions, 2024 to 2026, in their own words. Our reading of what Derek Lowe has said — not written or endorsed by them.

  1. Our readingThe hypotheses AI 'co-scientist' systems are producing are ones the literature already contains, and the publicity around them does not say so.

    Their wordsI think it’s good that the system identified the mechanisms that it did, but I don’t think that the paper does enough to point out that none of these ideas are without precedent - in some cases, a lot of precedent.

    ↗Evaluating "Co-Scientist", a New AI Science Systemweb.archive.org

  2. 19 months earlier

  3. Our readingThere are few clear-cut cases in biopharma today where buying an AI system is obviously worth it.

    Their wordsAt the moment, there are in fact not very many slam-dunk gotta-get-me-some-of-that-AI decisions out there in our business.

    ↗Half-Full and Half-Empty AI Glassesweb.archive.org

  4. 4 weeks earlier

  5. Our readingHow much machine learning can help a biological problem is set by the quality of the data around that problem, not by the algorithms.

    Their wordsI predict (with confidence bordering on arrogance) that whatever machine-learning progress can be made will correlate extremely closely with the quality of the data around any given problem

    ↗AI and Biologyweb.archive.org 1st of 4 in this piece

  7. Our readingMost proteins have no known function, so there is no pile of solid data for a model to learn most of cell biology from.

    Their wordsMost proteins are essentially unannotated, with their functions completely or mostly unknown.

    ↗AI and Biologyweb.archive.org 2nd of 4 in this piece

  9. Our readingBiology keeps turning up whole classes of things nobody knew existed, and that is what defeats attempts to model it.

    Their wordsThe "unknown unknown" problem makes hash (in retrospect) of many attempts at biological theorizing and modeling, and I think that this problem will still accrue to ML models until we know an awful lot more.

    ↗AI and Biologyweb.archive.org 3rd of 4 in this piece

  11. Our readingAI will not produce a big breakthrough in biological understanding any time soon, because biology has nothing like the clean, complete, high-quality data that made protein-structure prediction work.

    Their wordsSo for these reasons, I (unfortunately) do not expect any big AI-driven breakthroughs in biological understanding any time soon.

    ↗AI and Biologyweb.archive.org 4th of 4 in this piece