Through the spectrum of artificial intelligence uses, we stand out.
THE Great opportunition of the inspiring AI On the horizon, experts agree, lies in the acceleration and transformation of scientific discovery and development. Fend by large masts of scientific data, IA Promises to generate new drugs to fight diseases, new agriculture to feed the world’s population and new materials to unlock green energy – all in a small fraction of traditional research time.
Technological companies as Microsoft And Google Make AI tools for science and collaborate with partners in fields like Drug Discovery. And the Nobel Prize in chemistry last year went to scientists using AI To predict and create proteins.
This month, Lila Sciences went public with its own ambitions to revolutionize science through the AI The start-up, which is based in Cambridge, in Massachusetts, had worked in secret for two years “to build a scientific superintelligence to resolve the greatest challenges of humanity”.
Based on an experienced team of scientists and $ 200 million in initial funding, Lila has developed an AI program formed on published and experimental data, as well as on the scientific process and reasoning. The start-up then leaves that AI software performs experiences in automated physical laboratories with a few scientists to help.
Already, in projects demonstrating technology, Lila AI has generated new antibodies to combat diseases and developed new materials to capture carbon from the atmosphere. Lila transformed these experiences into physical results in her laboratory in a few months, a process that would most likely take years with conventional research.
Experiments like Lila have convinced many scientists that AI will soon make the cycle of experience-test hypothesis faster than ever. In some cases, AI could even go beyond human imagination with inventions, turbocharged progress.
“The AI will feed the next revolution in this most precious thing that humans have ever fallen – the scientific method,” said Geoffrey von Maltzahn, general manager of Lila, who has a doctorate. In biomedical and medical physical engineering of the Massachusetts Institute of Technology.
The thrust to reinvent the scientific discovery process is based on the power of the generative AI, which burst into public awareness of the introduction of the Openai chatpt just over two years ago. The new technology is trained on internet data and can answer questions, write reports and compose emails with the mastery of humanity.
The new AI breed has sparked a trade in commercial arms and apparently unlimited expenses by technological companies, including OPENAI, Microsoft and Google.
(The New York Times has heard OPENAI and Microsoft, who have formed a partnership, accusing them of copyright violation concerning news content linked to AI systems. Openai and Microsoft have denied these complaints.)
Lila has adopted a science -oriented approach to train her generative AI, nourishing computer research articles, documented experiences and data from her life science and rapid -growing materials laboratory. This, according to the Lila team, will give technology both the depth of science and large -scale capacities, reflecting the way chatbots can write poetry and computer code.
However, Lila and any business working to break the “scientific superintendent” will face major challenges, according to scientists. Although AI already is revolutionizing certain areas, including the discovery of drugs, it is not clear if technology is only a powerful tool or on a path to match or exceed all human capacities.
Since Lila operates in secret, external scientists have not been able to assess her work and, they add, the early progress of science does not guarantee success, because unforeseen obstacles often arise.
“More power for them, if they can do it,” said David Baker, a biochemist and director of the Protein design institute at Washington University. “It seems beyond everything I know in scientific discovery.”
Dr. Baker, who shared the Nobel Prize in chemistry last year, said he considered AI more as a tool.
Lila was designed inside Lighthouse pioneerAn investor and a prolific creator of biotechnology companies, including the COVVI-19 Moderna vaccine manufacturer. The flagship product conducts scientific research, focusing on the place where breakthroughs are probably in a few years and could prove to trade in trade, said Noubar Afeyan, founder and managing director of Flagship.
“So not only do we care about the idea, but we care about the speed of the idea,” said Dr. Afeyan.
Lila results from the merger of two first projects of the company of AI to the flagship product, one focused on new materials and the other on biology. The two groups tried to solve similar problems and recruit the same people, so they combined the forces, said Molly Gibson, a computer biologist and co-founder of Lila.
The Lila team has carried out five projects to demonstrate the capacities of its AI, a powerful version of one of an increasing number of sophisticated assistants called agents. In each case, scientists – who generally had no specialty in matters – typed a request for what they wanted the AI program to be completed. After refining the request, the scientists, by working with AI as a partner, led experiences and tested the results – again and again, regularly established themselves on the desired target.
One of these projects has found a new catalyst for the production of green hydrogen, which consists in using electricity to divide water into hydrogen and oxygen. The AI has been indicated that the catalyst should be abundant or easy to produce, unlike the iridium, the current commercial standard. With the help of AI, the two scientists have found a new catalyst in four months – a process that could generally take years.
This success helped persuade John Gregoire, an eminent researcher in new materials for clean energy, leaves the California Institute of Technology last year to join Lila as research manager on physical sciences.
George Church, a Harvard geneticist known for his pioneering research in the sequencing of the genome and the synthesis of DNA which co -founded dozens of companies, has also recently joined as a Lila chief scientist.
“I think science is a very good subject for AI,” said Dr. Church. Science focuses on specific areas of knowledge, where truth and precision can be tested and measured, he added. This makes AI in science less prone to wandering and erroneous responses, called hallucinations, sometimes created by chatbots.
The first projects are still far from the products ready for the market. Lila will now work with partners to market the emerging ideas of her laboratory.
Lila extends her laboratory space in a six -story flagship building in Cambridge, alongside the Charles river. Over the next two years, known as Lila, he plans to move to a separate building, to add tens of thousands of square feet of laboratory space and offices open in San Francisco and London.
A recent day, trays carrying 96 wells of DNA samples are mounted on magnetic tracks, quickly changing the directions for delivery to different laboratory stations, in part that the AI has suggested. Technology seemed to improvise when she has executed experimental steps in the pursuit of new proteins, gene editors or metabolic paths.
In another part of the laboratory, scientists have monitored high -tech machines used to create, measure and analyze nanoparticles of new personalized materials.
The activity on the laboratory floor was guided by a collaboration of white -coated scientists, automated equipment and invisible software. Each measurement, each experience, each incrementally success and failure has been captured digitally and fed in Lila AI so that it learns continuously, becomes more intelligent and in fact more alone.
“Our goal is really to give AI an access to manage the scientific method – to offer new ideas and to go to the laboratory and to test these ideas,” said Dr. Gibson.