Anthropic’s IPO filing on Monday dropped the revolutionary rhetoric about biology and drug discovery that CEO Dario Amodei has pushed for years.
Two years ago, Amodei suggested AI could deliver a thousand years of biological progress in just five to 10 years. In August, he told X that the company was ramping up efforts quickly in medicine. He predicted most human diseases would be cured in the next decade and promised a flood of AI-accelerated drugs. The company also claimed in September that Claude had discovered a “Crispr-like” enzyme.
The new filing still claims AI can “profoundly accelerate the rate of progress in biology and medicine” and compress decades of work into years. However, it is far more reserved about turning that potential into reality.
Anthropic states its teams “need first-hand experience doing scientific research and drug discovery.” It acknowledges that life sciences work requires activities outside its history as a software and research organisation. This means handling hazardous materials, running laboratory equipment, and following strict regulations.
The company also softened its language on the enzyme discovery. It now says the find “might have the potential to be developed into a novel biotechnology tool, perhaps for gene editing.” Researchers previously told WIRED they were doubtful of the enzyme’s value. One scientist studying those enzymes suspected Claude had accessed his private data. Anthropic said neither the model nor its molecular biology team had access to that material.
The IPO filing shows Anthropic no longer sees itself solely as an outside provider of AI tools to biotech firms. It treats biotech discovery as a core part of its business model. Like all biotech companies, it faces long timelines. Any AI-assisted discoveries must be tested in test tubes, then in animals, and eventually in people. Even if AI helps identify which new drugs or targets work in humans versus those that fail, years of testing remain to prove safety and efficacy.
Anthropic recognises these hurdles in the filing. “Any drug candidates or other discoveries resulting from these efforts would face the lengthy, costly, and uncertain development and regulatory pathways characteristic of the life sciences industry, and may never generate revenue,” the document states. Amodei argued in his August X post for streamlining regulations for the upcoming wave of AI-generated drugs.
US Security and Exchange Commission filings are generally more cautious than statements made by companies or their executives due to the legal risks of overstating claims to investors. The IPO makes clear the company believes its technology will drive revolutionary advancements.
As Anthropic pushes further into biological discovery, it faces the same problem as other biotech firms: proving worth in trials that cannot be hastened simply by adding computers.
What it means
People making things in biology or medicine should expect slower timelines. Anthropic is moving from writing code to running physical experiments. This shift brings regulatory hurdles and safety checks that software alone cannot bypass. The promise of rapid cures remains theoretical until the company proves its models work in real-world lab conditions.
