Former testicular cancer patient whose father didn't live to see a life-saving drug: Anthropic CEO predicts conquering cancer, Alzheimer's disease within a decade, and even doubling human lifespan.

Beating AI News Flash: Anthropic CEO Dario Amodei made a bold prediction during the company's IPO roadshow: cancer, Alzheimer's disease, heart disease, and mental illness could become history in less than 10 years. He believes that if AI risks can be properly controlled, humanity may usher in a "miracle era" that solves problems thousands of years old.

As early as 2024, Amodei proposed the "compressed 21st century" in his essay "Machines of Loving Grace": with the help of AI more capable than top scientists, medical progress that would originally take 50 to 100 years could be concentrated into 5 to 10 years. At the time, he even envisioned that AI could reduce cancer incidence and mortality by more than 95%, prevent Alzheimer's disease, and extend average human life expectancy from about 75 years to 150 years.

Amodei has personal experience with disease and medical progress. He once had early-stage testicular cancer and was successfully treated, while his father died of hepatitis C. A few years after his father's death, a new generation of antiviral drugs came out, achieving a cure rate of about 95%. In September this year, he mentioned this experience again, saying he deeply understands what it means to patients when medical breakthroughs arrive a few years earlier.

Anthropic has already begun participating directly in new drug development. In October, the company disclosed collaborations with six biomedical institutions including Bristol Myers Squibb, and also acquired Mithril Medicine, a drug development company with only one full-time employee.

Mithril's founder used Claude to manage experiments, analyze data, and coordinate 11 external laboratories, advancing a potential therapy for a rare muscular atrophy disease to animal testing in less than 18 months. Some treated mice achieved about twice the grip strength and muscle strength of the untreated group, and their survival also improved. Anthropic expects that if subsequent experiments go smoothly, the drug could enter human clinical trials as early as the end of 2028.

But while AI can accelerate drug design and analyze experimental data, it cannot directly shorten the time required for the human body to produce therapeutic effects or develop long-term side effects. Especially for chronic diseases such as Alzheimer's disease, researchers may need to track patients for years to confirm whether a new drug is truly effective.

Amodei has also acknowledged that clinical trials and regulatory approval are the main obstacles to accelerating medicine. In an article in June this year, he noted that new drugs usually take about 7 to 8 years to pass regulatory approval in the United States and Europe. He advocates using better AI simulation, toxicity prediction, and biomarkers to shorten development cycles while retaining necessary safety verification.

At the end of the IPO roadshow, Amodei also emphasized that only companies capable of controlling AI risks and earning public trust are qualified to continue developing and applying this technology. He linked the vision of curing major diseases with Anthropic's long-standing emphasis on AI safety.