Something happened a couple weeks ago that will be remembered as a pivotal moment in the AI revolution.

Anthropic revealed it’s been quietly running a biology lab.

Yes, the same Anthropic behind Claude, the AI chatbot that competes with ChatGPT, now has scientists working in a physical laboratory in the Bay Area.

And they’ve already found something.

Anthropic recently turned hundreds of Claude AI agents loose on a massive database of DNA sequences. Their mission was relatively simple: Look for something interesting.

For 21 hours, roughly 950 AI agents combed through the data, investigating unusual families of proteins and looking for patterns. Eventually, one of the agents spotted something strange… a repeating pattern of DNA next to an unusual enzyme.

Scientists took a closer look and realized Claude appeared to have identified a previously uncharacterized enzyme system with features reminiscent of CRISPR, the revolutionary gene-editing technology that transformed biotechnology.

Now, let’s not get ahead of ourselves. Anthropic doesn’t yet know exactly what this enzyme system does. The research is early, and the findings have been released as a preprint rather than peer-reviewed research.

But that’s not what grabbed my attention.

AI’s gone from language professor to mad scientist.

From Igor to Frankenstein

Until now, most people have experienced AI as an assistant. We ask it questions. It writes emails. It analyzes spreadsheets. It helps computer programmers write code.

Anthropic is trying something much more ambitious: graduate from assistant to senior scientist.

Earlier this year, it assembled a team of scientists to determine whether AI agents could accelerate biological discovery. The company built its own wet lab so Claude could generate hypotheses, humans could test them in the physical world, and those experimental results could then feed back into the AI.

And Anthropic isn’t stopping there. Last month, it launched a program giving qualified life-sciences organizations expanded access to its most advanced models for drug discovery, research biology, clinical development, and manufacturing.

Think about where this could eventually lead.

Biology may be the ultimate big-data problem. The human body contains enormous amounts of genetic, molecular, and clinical information. Finding meaningful relationships buried inside that data can take researchers years.

AI can search it at a scale no human possibly could.

And that’s why the convergence of AI and biology could become one of the most important investment trends of the next decade.

Cancer, Proteins, and Whispers in the Blood

Fortunately, Anthropic isn’t the only company working at this intersection. A new generation of publicly traded companies is emerging around AI, genomics, proteins, and precision medicine.

One that immediately caught my attention is Caris Life Sciences (CAI).

Imagine two people are diagnosed with the same type of cancer. On the surface, their cancers may look similar. But at the molecular level, they can be very different, which means the drug that works for one person may not work for the other.

Caris is using AI to help solve that problem.

The company analyzes cancer tumors and compares what it finds against an enormous database built from more than one million patients. AI can then search through all that information for patterns that could help doctors better understand a patient’s cancer and determine which treatments are most likely to work.

In other words, instead of treating cancer based primarily on where it appears in the body, we’re moving toward treating your specific cancer based on what makes it tick.

That’s personalized medicine. And AI could put it on steroids.

Caris only went public last year, but this isn’t some tiny biotech with nothing but a promising idea. Revenue jumped 45% last quarter to $264 million, and management expects to top $1 billion for the year.

Then there’s Generate Biomedicines (GENB), which just went public in February.

This one sounds like science fiction.

Most traditional drug discovery starts by searching for a molecule or protein that might affect a disease. Researchers can spend years testing thousands of possibilities hoping to find something that works.

Generate wants to flip that process around.

Its AI starts with the problem and attempts to design the protein needed to solve it.

Think about ChatGPT. You give it instructions and it generates new text. An image generator can create a picture that never existed before.

Generate is attempting to do something similar with medicine.

Tell the AI what you want a protein to do, and the technology can potentially design one specifically for that job. That could eventually mean creating medicines rather than simply searching for them.

We’re still early here. Generate is a clinical-stage company, which makes it much more speculative than Caris. But it represents exactly the kind of breakthrough that gets me excited about where AI is heading.

And then there’s Alamar Biosciences (ALMR), which went public just six months ago.

Here’s the easiest way to think about Alamar.

Our blood is constantly carrying clues about what’s happening inside our bodies. Proteins can provide early signals of cancer, Alzheimer’s, inflammation, and other diseases. The problem is that some of those signals are so faint that today’s technology has trouble detecting them.

Alamar is building technology designed to find those whispers.

Its systems can measure proteins present in extraordinarily tiny amounts, potentially allowing researchers to spot biological changes much earlier than they could before.

Now combine that with AI.

The more information we can pull from a simple blood sample, the more data AI has to analyze. Eventually, that could help doctors identify disease before symptoms appear; when there’s a much better chance of doing something about it.

And once again, there’s a real business beginning to emerge. Alamar’s revenue surged 82% last quarter to $29.4 million, and management expects roughly 59% growth this year.

Bigger Than Better Emails

Three companies. Three very different approaches.

Caris wants to use AI to figure out which cancer treatment is right for you.

Generate wants AI to help design entirely new medicines.

Alamar wants to uncover signs of disease hiding in your blood.

That’s when this trend starts to gain real traction.

We’re not just talking about AI writing emails faster anymore.

We’re talking about using it to understand cancer, create drugs, and potentially detect diseases before we even know we’re sick.

P.S. Claude may be learning to do science, but the company behind it is still off-limits to most investors. Anthropic, like OpenAI, remains private.

But in my new presentation, I show you how to get exposure to both today through already-public companies sitting on big stakes in them, before either one rings the opening bell. Watch it here.