Following a three-decade career at the helm of some of Silicon Valley’s most powerful companies—cofounding LinkedIn and sitting on the boards of PayPal and OpenAI—Reid Hoffman recently turned his attention to health care.
Hoffman’s startup, Manas AI, is building an AI engine that aims to fast-track the traditionally slow process of drug discovery for various cancers. Inspired by a dinner with renowned cancer physician Siddhartha Mukherjee, the company’s cofounder and CEO, its mission statement is to “shift drug discovery from a decade-long process to one that takes a few years.”
But Hoffman’s enthusiasm for generative AI, in particular, stretches far beyond novel drug targets and small molecules. He believes that frontier models—the most advanced, large-scale AI models currently available from companies like OpenAI and Anthropic—should be a cornerstone of health care itself.
“If as a doctor, you’re not using one or more frontier models as a second opinion, my belief is you’re bordering on committing malpractice,” Hoffman said, speaking at WIRED Health in London on April 16. “These AI systems, even though many of them are not specifically trained for medicine, have ingested trillion-plus words of information. As a second opinion, it is bringing superpowers that no human being has.”
Such comments will undoubtedly rattle many doctors. Earlier this year, a major study concluded that large language models present risks to members of the general public seeking medical advice due to their propensity for providing inaccurate and changeable information.
Hoffman’s argument is that rather than outsourcing critical thinking capabilities to AI models, people should use them as an additional source of information, one that he believes could prevent misdiagnosis. He claims to personally use frontier models as a second opinion for issues relating to his own health and insists that his personal concierge doctors do so as well.
“You could very well go, ‘No, I think you’re wrong, I think it’s this,’” he told the WIRED health audience. “But if you’re not using this as a second opinion, you’re making a mistake, both as a doctor and as a patient.”
With the UK’s National Health Service buckling under the strain of extensive waiting lists and workforce challenges, including a chronic shortage of family doctors, Hoffman believes there’s an increasingly pressing need for a large language model that could act as a free medical assistant on every smartphone. He suggests it could also serve as a form of early triage for appointments with human doctors.
“We just don’t have enough doctors, most people don’t have access, and when you think about, ‘How should the NHS be redesigned?’ everyone should be interacting with this medical assistant,” he said.
While he has a conflict of interest as an entrepreneur working in drug discovery, Hoffman is also keen to see AI play a wider role in assisting the FDA and other regulators in assessing emerging medicines, as well as accelerating the availability of particularly promising drugs to patients.
“As a Silicon Valley person, I would love to get to a point where the FDA was also running tests with biological models, going, ‘Oh, we should fast-track this one, because the likelihood of negative consequences is lower,’” he said. “Do I think that’s anytime soon? Unfortunately, no.”
As for Manas AI, human judgment still plays a key role in the company’s decisions regarding which targets to pursue. Mukherjee closely reviews their AI engine’s proposals, Hoffman says, and sifts the genuinely interesting candidates from the “bonkers stupid.”
While the company’s initial focus is on cancer, Hoffman believes that the potential of AI discovery engines is much broader, enabling the identification of drug candidates for chronic but also extremely rare diseases that have not traditionally been as economical for pharmaceutical companies to research.
“I think in 10 years, every major disease will have target molecules that could at least make a serious difference,” Hoffman said.
