AI News

Anthropic CEO says cancer will become history within a decade; medical community questions his IPO narrative

36Kr · AIUpdated Nov 13 21:44

AI summary

Anthropic CEO Dario Amodei said in an IPO roadshow video that if AI is used well, cancer, Alzheimer's disease, heart disease and mental illness will become relics of the past in less than ten years. Former German Health Minister Karl Lauterbach said doing so in just 10 years is impossible, while Vertex Executive Chairman Jeff Leiden said his claim is nonsense and noted that the company's collaboration with about 50 AI companies has not produced a single effective drug. Anthropic launched its IPO roadshow on November 10, planning to issue 825 million Class A ordinary shares at an expected price of $70 to $78 per share, and underwriters are considering raising the offering price.

Why read this

AI score 60

A review of Amodei's ten-year cancer cure prediction during the IPO roadshow and the medical community's rebuttal, presenting the divergence between capital narrative and scientific prudence.

Topics

Tags

Body · AI translation

If you are an investor, then you should have been very excited during the week that just ended.

Because Anthropic's head, Amodei, predicted during the IPO roadshow that AI will make cancer a thing of the past within 10 years.

If you care about the pharmaceutical industry, then you should also have been very excited during the week that just ended.

Because Amodei set a specific timeline for solving cancer for humanity.

Anthropic's head makes bold claims again

"You know, if we look at cancer, Alzheimer's, heart disease, mental illness and other problems today, I think that in less than ten years, these will all become relics of the past."

In the company's IPO roadshow video, Anthropic CEO Dario Amodei said that if humanity can make good use of the power of AI, then widespread problems that have long tormented humanity, such as cancer and mental illness, could all become history in just 10 years.

In fact.

This is not the first time this CEO has linked "the future of humanity" with AI. As early as an interview with "60 Minutes," he predicted that AI would find treatments for most cancers, and that human life expectancy would multiply with the continuous development of AI technology.

The famous "compressed 21st century" theory was also proposed by him on that program ↓

If AI can work with existing top scientists and researchers, then the pace of human medical progress is expected to increase tenfold. In this context, the medical advances humanity could achieve throughout the 21st century would also appear earlier.

But...

To understand a person, we cannot just look at what he says, but also observe what he does.

On November 10, Anthropic officially announced the launch of its proposed initial public offering (IPO) roadshow, and plans to issue 825 million Class A ordinary shares. It is worth noting that although Company A's IPO price is expected to be between $70.00 and $78.00 per share, due to excessive investor enthusiasm after the roadshow began, underwriters are currently considering raising the offering price.

This also means that within the next few days, Anthropic's IPO valuation target is expected to be further pushed up to $1.5 trillion or even higher.

Pharma leaders pour cold water on it one after another

Now, Amodei's bold claims have not only shaken the financial markets, but even normally low-profile medical leaders have come forward to respond.

A few examples ↓

Former German Health Minister Karl Lauterbach said on social media: "Using AI to cure cancer, Alzheimer's and mental illness may indeed bring benefits, but doing so in just 10 years is impossible."

"We don't know whether people like Dario Amodei know nothing about medicine or are deliberately misleading the public."

Melinda B. Chu, an MD from Saint Louis University, also complained after learning all this that Amodei's series of remarks were nothing more than an attempt to add heat to his IPO. Before there is evidence that Anthropic has made outstanding contributions in areas such as "curing cancer," this news is merely hype.

By contrast, the remarks of Jeff Leiden, executive chairman of Vertex Pharmaceuticals (a veteran biotech company founded in 1989), seemed more "direct." In his view, Amodei was completely "talking nonsense." Because Vertex had cooperated with about 50 AI companies, hoping to use artificial intelligence technology to sort through a data package containing tens of thousands of drug compounds.

Unfortunately, they did not develop an effective drug.

"It simply doesn't work on its own."

"Any company claiming it can do this—at least in our experience, we examined most of them—is not telling the truth." Jeff Leiden said so in relevant reports.

In fact, although AI drug discovery is highly hyped. To date, no drug "fully designed by AI" has received FDA approval for marketing. That is why executives at biopharma giants such as AstraZeneca say: stay in this industry long enough, and people learn humility. Disease is complex, and biology even more so.

But still, as the saying goes: "Saying is one thing, doing is another."

Although AstraZeneca executives remain conservative about AI drug discovery, they still signed a contract worth $40 million upfront with Insitro (an AI biotech company), in an attempt to discover new weight-loss drugs and heart disease drugs. In addition, AstraZeneca has formally established a partnership with Oracle, hoping to optimize the recruitment process for clinical volunteers.

On the data level, McKinsey said that in 2025 alone, venture capital and private equity around the world invested $8.4 billion in the "AI drug discovery" track. Compared with about $4 billion two years ago, this is a running acceleration into the sector.

In addition, CVSource data from Zhongjia Chuan also pointed out that in the first nine months of 2026, 57 investments were completed in the domestic AI pharmaceutical track, with a disclosed total financing amount of 15.765 billion yuan. This figure has already exceeded the total for 2023-2025.

Returning to Anthropic itself, the real distance between it and AI drug discovery may be much greater than Amodei described. After all, the business focus the company presents externally is still concentrated on programming, not drug R&D itself.

In other words, even if AI theoretically has the potential to accelerate medical discovery, from algorithms to clinical practice, from compound screening to approval and marketing, what lies in between is still a long trial cycle, strict regulatory review and an extremely high failure rate. These cannot be bypassed simply by improving model capabilities.

What is more intriguing is that Amodei's optimistic narrative happened to appear at a key moment in the IPO roadshow.

When a company needs to tell a story imaginative enough in the capital market, binding AI to "conquering terminal diseases" is undoubtedly the option most likely to ignite investor sentiment. But the enthusiasm of the capital market and the prudence of the scientific community are inherently two different evaluation systems. The logic of the former is expectations and valuation, while the logic of the latter is evidence and reproducibility. The collective rebuttal from medical leaders is essentially vigilance against this narrative crossing boundaries.

On the contrary, from target discovery to clinical trial design, from patient recruitment to data cleaning, AI is penetrating every link of drug R&D in a more pragmatic and fragmented way. It is just that these advances are often dull, slow, and difficult to compress into an exciting slogan. After all, real medical progress never has a specific timetable; it advances slowly through day-after-day verification and trial and error.

References:

Tianyancha

Cailian Press: Anthropic CEO's shocking prediction: AI may conquer cancer, heart disease and mental illness within ten years!

Wallstreetcn: Pharma CEOs collectively pour cold water: the promise that AI will "cure cancer in ten years" can be made by Silicon Valley, but biology cannot do it

Jiemian News: Anthropic launches proposed initial public offering roadshow

STAR Market Daily: Investors flock in! Anthropic's IPO subscription multiple reaches about 6 times

Wallstreetcn: Report: Anthropic plans to raise its offering price, seeking a $1.5 trillion valuation, with fundraising scale possibly exceeding $100 billion

This article comes from the WeChat public account "mawen011" (ID: hlw0823), author: Internet Those Things, authorized for publication by 36Kr.

Source

36Kr · AI