Lee Kai-Fu says artificial intelligence will speed up drug discovery
This article is an on-site version of our #techFT newsletter and is generally only available with a Premium subscription. Are you currently a Standard subscriber? Click here to upgrade your subscription to Premium. Not yet subscribed? You can subscribe to a Premium subscription here
The pandemic has accelerated technology deployment in public health and medicine. Everything from contact tracing apps to services offering remote medical appointments. But nowhere has the role of technology – particularly artificial intelligence – been more hotly debated than in the field of drug discovery.
Proponents point to the pandemic as proof of its value in healthcare, even as critics dismiss its use in drug discovery as “hype,” reports Hannah Kuchler.
In January 2020, scientists at pharmaceutical company BenevolentAI used artificial intelligence algorithms to scour 50 minutes of medical journals to search for approved drugs that could be repurposed to treat disease.
Scientists and the algorithm narrowed the search to baricitinib, used to treat rheumatoid arthritis, all within the space of four days. The drug Eli Lilly tackled both the virus and the body’s inflammatory response. The event marked AI’s first discovery of a drug, already widely used, that could be redeployed.
TechFT caught up with Lee Kai-Fu, an investor and author of AI 2041: ten visions for our futureto talk about the role of technology in drug discovery.
HEY : What role does AI play in drug discovery?
LKF: AI can help accelerate drug discovery in three phases of drug development. The first phase uses AI to reduce the number of drug candidates. Instead of a scientist filtering a drug out of 10,000 candidates, the AI will do so based on input provided by the scientist. This process combines AI and human expertise to select candidates with the best chance of success in clinical trials.
The second phase uses AI during the experimentation phase, where robots can automate much of the lab work. The robots reproduce simple and routine manual procedures such as opening test tubes, mixing liquids, introducing chemical agents, growing the culture, observing the reaction and producing the result. These are routine and reproducible procedures for robots, thus accelerating the second phase of clinical drug development.
The third phase is to use AI to speed up processes at the clinical trial stage, for example by helping pharmaceutical companies match patients to the clinical trial. The benefit of these three combinations is to reduce the cost of drug discovery, lowering the bar for pharmaceutical companies to develop cures for rare diseases that were uneconomical to target.
HEY : There is a lot of skepticism in the medical establishment about the role of AI in medicine. What will convince them of the future you are designing?
LKF: I am very aware of a conscious disconnect between the medical community and the AI community. There are plenty of AI people with starry dreams that they could change the world if only their software were adopted everywhere – who think that’s not happening just because of a different way of thinking [in the medical profession]. But this problem will not arise in drug discovery because the interests are aligned. We are not proposing a new method of drug discovery. In each of these three phases, everything will be done to the precision and satisfaction of humans, within a framework that already exists.
HEY : What are the risks of deploying AI in drug discovery?
LKF: There’s a chance that the same tools used to discover drugs that cure disease could be used to invent toxins to hurt people with precision. A possible mitigation is to not make this source open to prevent it from falling into the wrong hands.
Learn more about this topic with Hannah Kuchler great read about AI’s goals of energizing the hunt for new drugs, which some critics dismiss as a pipe dream. At Anjana Ahuja’s article the warning about the dangers of AI that could be used to develop bioweapons provides a useful caveat to pharmaceutical and AI companies.
The Internet of (four) things
1. South Korean fintech challenger targets Southeast Asia
Viva Republica, one of South Korea’s most valuable fintech startups, is seeking to raise up to $1 billion from international investors. Toss offers money transfer and debit card services in South Korea and Vietnam. It is now entering new markets in Southeast Asia, where it will take on Singapore’s Grab and Indonesia’s GoTo, which analysts say have the incumbent advantage, having built a robust fintech ecosystem on the strong carpooling database.
2. Amazon faces crucial vote on unionization
Workers at an Amazon Alabama warehouse voted on whether they wanted to unionize, a step that could mark the company’s first unionization in the country. Amazon, the second largest employer in the United States, after Walmart, has an unmatched record for crushing union organizing efforts in the United States. Dave Lee reports that a “yes” vote would spell trouble for Amazon, just as it is squeezed by rising labor costs and supply chain shortages. Ballots are to be counted next week.
3. Big Tech’s expanding land grab
Lex dug through the annual reports of major tech companies to trace the large real estate investments made after accumulating large piles of cash. In the latest round of annual reports, Amazon listed its global properties as close to 35 million square feet, plus an additional 570 million square feet of leased space. It values its land and buildings at $81 billion, up from $32 billion at the end of 2018. Meta totals the value of its land and buildings at $24 billion in 2021, up from $8 billion in 2018.
4. EU set to unveil landmark tech giant law
The EU is set to unveil legislation to limit Big Tech’s market power as early as Thursday. The Digital Markets Act will impact internet businesses with at least 45,000 active users and introduce new controls such as tighter restrictions on how businesses handle personal data.
Dyson’s new and improved Airwrap multi-styler hair tool will hit UK store shelves next month – an upgrade to the sold-out original 2018 model. Airwrap can dry, curl, straighten and shape hair – without the same heat damage as traditional curling irons or straighteners. The Airwrap uses air currents to shape strands of hair, and the new model has attachments that make sprucing up your barnet quicker and easier. The complete Dyson Air envelope kit will set you back £479.99.