Real Alpha
Watch modeSubscribed

The MAD Podcast · 2 min · Oct 1, 2026

Interpretability revealed a new Alzheimer's biomarker

Goodfire traced an Alzheimer's model's blood-draw predictions to DNA fragment length, a signal its developers had not identified. The case shows how interpretability can uncover scientific findings, not just debug behavior.

Eric Ho

Eric Ho is cofounder and CEO of Goodfire.

Co-founded Goodfire, the interpretability research lab, and previously scaled RippleMatch to over $10M in ARR.

  • Matt Turck, Host · General Partner, FirstMark Capital

    Publishes the annual MAD landscape of data and AI; sold TripleHop to Oracle.

About this segment

Goodfire traced an Alzheimer's model's blood-draw predictions to DNA fragment length, a signal its developers had not identified. The case shows how interpretability can uncover scientific findings, not just debug behavior.

From “AI Models Are Now Hiding Their Cheating | Goodfire”, Oct 1, 2026. Watch the full episode

“we were able to reverse engineer the computations of their model and find a new Alzheimer's biomarker, which ended up just being a fragmentomic biomarker.”
Eric Hoat 1:02:25 · The MAD Podcast1:02:25

Takeaways

  1. The model diagnosed Alzheimer's from a blood draw.
  2. Interpretability linked its predictions to DNA fragment length.
  3. The team says the model's developers had not identified the biomarker.

In this episode

Full episode
  1. 8:12Impossible tasks and weak sandboxes can turn agents rogue3:33
  2. 25:34A model planned a hack to evade its chain-of-thought monitor5:11
  3. 53:22Interpretability can monitor, diagnose, and shape model behavior6:14
  4. 4:09Open models encode reward hacking in their activations4:08
  5. 44:12A simple activation contrast can detect reward hacking3:16
  6. 49:13Cheap internal monitors can triage costly model reviews4:14

  7. 1:01:27Interpretability revealed a new Alzheimer's biomarker2:12

  8. 21:22Heavy reinforcement learning puts reward hacking front and center4:17Up next

Transcript

Transcribed by machine (Whisper) from the episode’s audio, so a word may be off here and there.

  1. 1:01:30

    ARC Institute, Mayo Clinic, Prima Mente. What's the use case there? The use case there is often quite similar. It can be a debugging use case where the model has learned some type of shortcut or some type of problematic behavior. And we can kind of go in and diagnose and remove what that behavior is. Or what's most exciting in life sciences, at least to me, is this concept of finding novel scientific knowledge within these models directly.

  2. 1:02:06

    So one example of this was the work that we did with Prima Mente, where they had a state-of-the-art Alzheimer's diagnostics model. They didn't know how it worked. They had no idea. It was just a black box. And I was able to diagnose, you know, new patients directly from a blood draw, whether or not they had Alzheimer's. So we were able to reverse engineer the computations of their model and find a new Alzheimer's biomarker, which ended up just being a fragmentomic biomarker.

  3. 1:02:41

    So like fragment length ended up being a strong predictor of Alzheimer's disease. and this was very surprising because they didn't know this going in and so we're able to do this type of unsupervised discovery directly in the weights of these models to find novel novel concepts i think this is going to become increasingly interesting across scientific domains especially in the life sciences but also i'd be so curious to try to figure out like Like, you know, when OpenAI's models are solving like, you know, novel mathematical problems, like what's going on inside their mind?

  4. 1:03:26

    There must be something happening there that is quite novel that we can uncover and reverse engineer and see what's happening.