Top AI Scientist "Gone Mad" and Executes His Nephew In Front Of Courthouse artwork

Top AI Scientist "Gone Mad" and Executes His Nephew In Front Of Courthouse

Rotten Mango

July 26, 2026

One prominent UFO researcher once said, if you start looking for patterns, you will find them. And people think one of the worst patterns to emerge in 2026 is the pattern of important scientists going missing or ending up dead.
Speakers: Stephanie Soo
**Stephanie Soo** (0:02)
One prominent UFO researcher once said, If you start looking for patterns, you're going to find them. And people think one of the worst patterns to emerge in 2026 is the pattern of important scientists going missing or ending up dead. Is it a fear-mongering conspiracy where netizens are connecting together a few dots and calling it a link? Or is it a real, legitimate theory? I mean, I can't be the one to tell you. However, it does seem that any scientists or researchers studying something important to the greater safety of humankind or the detriment of humankind, anyone, scientist, researcher involved in some sort of death, it's going to be investigated by some curious netizens.
So Chen, a Chinese researcher based in the United States, he has the exact type of resume where if something happens to him, it makes you want to throw coincidence out the window and sign up for a front row seat at ConspiracyCon.
Chen has a research role at CASIA, which is basically China's flagship AI research institution right now. Although they were first formed, I will say, to help China build nuclear missiles and satellites, their first program was actually called Two Bombs, One Satellite, which is like...

**SPEAKER_2** (1:16)
Two bomb?

**Stephanie Soo** (1:17)
Yeah, Two Bombs, One Satellite.

**SPEAKER_2** (1:19)
Two bomb, one satellite, okay.

**Stephanie Soo** (1:21)
And CASIA's signature specialty is actually biometrics, which is how they teach computers. How to identify humans by face, by their iris, and even their gait, how they walk. They had this whole thing where they could identify you. They don't even know you, they don't know what your face looks like, but they could identify you from a mile away on CCTV footage if they follow certain gait patterns. Yeah. They have a free data set of people's faces that have become one of the most widely used training set in the entire world for facial recognition. He was a postdoctoral researcher in the bioinformatics division at Tsinghua University. What does that even mean? What is bioinformatics? It sounds kind of like cute, but it's not cute at all. It's a very interesting role. It's not just data, it's not just AI, machine learning. You have to specialize in all of that, but you also have to specialize in molecular and cell biology. Researchers in the bioinformatics division at Tsinghua are working on projects like the Human Cell Atlas, which is actually a giant international project, like a global initiative. It's to catalog every cell type, describe its molecular fingerprint, if you will, and notate where it lives and what it does. Basically, Google Maps for the human body, which the human body is made up of roughly 37 trillion cells. So this is going to take a minute. The goal would be to give every biologist a shared reference. So if they find a weird cell in their own experiment, they can look it up against the master map. It would also make it easier to find out which cells changed, how, you know, which could help find cell signatures that flag early cancer before it's dangerous. But if you're not working on the single cell atlas, they could be in any number of different AI and biology research programs. The whole idea is LLM, large language models, so ChatGPT, cloud, but instead of being trained on human language, they're going to be trained on the human body.
Biology, like a language, create a foundational model, a cloud for cells, ChatGPT for cells. They train on cells and patterns of which genes turn on together, in which cell types, under which conditions. The ultimate goal, they said, is to have a digital patient. So everybody has a patient profile, and now they have one for AI, and you run experiments and medications through the digital patient before the real patient, which sounds like it could also save lives, but also sounds really fucking terrifying at the same time, okay? But after those research positions in China, Chen gets invited to come to the United States. He doesn't try to immigrate. He gets an invitation. It's like a Hogwarts invitation in the mail, for a research role for Cold Spring Harbor Laboratory, which has been dubbed the birthplace of molecular biology. So there is that. They've been around since 1890 They're a world-renowned, non-profit, private research and education institution. The lab today focuses on five distinct areas, cancer, neuroscience, genomics, plant biology, and quantitative biology.
They do a lot of important work, but also they were the epicenter for the American eugenics movement at one point.
So the lab did preserve all its records for teaching and research while openly discussing it. So I guess they're trying to be transparent about one of the darkest chapters of American history and science. But I mean, who's to say, right?

57 more minutes of transcript below

Feed this to your agent

Try it now — copy, paste, done:

curl -H "x-api-key: pt_demo" \
  https://spoken.md/transcripts/1000651996090

Works with Claude, ChatGPT, Cursor, and any agent that makes HTTP calls.

From $0.10 per transcript. No subscription. Credits never expire.

Using your own key:

curl -H "x-api-key: YOUR_KEY" \
  https://spoken.md/transcripts/1000778293594