Konrad Körding | Helping Human Scientists Do Better Science @ Vision Weekend Puerto Rico 2026

Why It Matters

By embedding rigorous self‑scrutiny into AI‑assisted workflows, the proposal could raise the overall quality of published research and mitigate the incentives that currently reward quantity over insight.

Key Takeaways

• Scientists often ask poorly defined, low‑impact research questions.
• AI tools tend to reinforce bad science without clear quality metrics.
• Introducing deliberate “friction” can help researchers spot logical fallacies.
• PlanYourScience.com offers a checklist‑driven interface to improve rigor.
• Publication pressure fuels volume over value, undermining scientific credibility.

Summary

At Vision Weekend Puerto Rico 2026, Konrad Körding argued that the biggest obstacle to better science is not a lack of data or computing power, but the way human researchers formulate and test questions.

He observed that many academics pose poorly defined, low‑impact questions and then use standard statistical tricks that generate false positives—citing the XKCD “jelly‑bean acne” meme as an illustration. AI‑driven tools such as Prism, Future House, and Sapio promise to automate discovery, yet Körding warns they inherit the same bias because they are trained on published literature, which includes both groundbreaking and junk papers.

Körding’s solution is to add “friction” rather than remove it. His prototype, PlanYourScience.com, forces users to articulate the precise gap they aim to fill and runs a checklist of common failure modes—over‑generalization, missing controls, and other fallacies he has catalogued from a century of bad science. He cites AlphaFold as a genuine success, but dismisses hype around AI‑generated drug candidates that merely re‑discover existing work.

If adopted, this friction‑based approach could curb the flood of low‑value publications, reshape peer‑review incentives, and create a more reliable feedback loop between researchers and AI. Körding invites the community to test the tool, suggesting that a systematic audit of scientific reasoning may be the missing piece for truly intelligent research assistants.

Original Description

This talk is part of Vision Weekend Puerto Rico 2026, a three-day gathering of researchers and builders in Old San Juan.

The program featured 30+ presentations, along with keynotes, lightning talks, office hours, and discussions focused on “AI for X” — exploring how AI is advancing fields like neurotech, biotech, energy, and beyond.

https://foresight.org/events/vision-weekend-puerto-rico-2026/

══════════════════════════════════════

About The Foresight Institute

The Foresight Institute is a research organization and non-profit that supports the beneficial development of high-impact technologies. Since our founding in 1986 on a vision of guiding powerful technologies, we have continued to evolve into a many-armed organization that focuses on several fields of science and technology that are too ambitious for legacy institutions to support. From molecular nanotechnology, to brain-computer interfaces, space exploration, cryptocommerce, and AI, Foresight gathers leading minds to advance research and accelerate progress toward flourishing futures.

We are entirely funded by your donations. If you enjoy what we do please consider donating through our donation page: https://foresight.org/donate/

Visit https://foresight.org, subscribe to our channel for more videos or join us here:

• Twitter: https://x.com/foresightinst

• Facebook: https://www.facebook.com/foresightinst

• LinkedIn: https://www.linkedin.com/company/foresight-institute