“Artificial intelligence models tend to make things up. For example, when asked to provide sources for their claims, AI chatbots will sometimes reply with a list of impressive-sounding papers that do not exist. These hallucinations can be frustrating, but, as the New York Times recently reported, they are also proving to be extremely useful during the early stages of scientific research. Just as AI models can invent plausible but imaginary literature, they can also hallucinate molecules that conform closely to the forms of nature but are entirely unknown to it. Scientists can reach into this bottomless well of creativity and pull out extraordinary discoveries,” notes The Doomslayer.
Artificial intelligence sometimes makes up information that appears to be factual, known as hallucinations. For example, artificial intelligence made up non-existent court decisions when an injury victim’s lawyers used it to find cases to cite in a court brief. And it made up false information that it included in patients’ medical summaries. A Google chatbot made up false claims that caused Google’s stock market value to shrink by $100 billion.
But as the New York Times explains, “In the universe of science,” “innovators are finding that A.I. hallucinations can be remarkably useful. The smart machines…are dreaming up riots of unrealities that help scientists track cancer, design drugs, invent medical devices, uncover weather phenomena and even win the Nobel Prize.”
“It’s actually giving scientists new ideas. It’s giving them the chance to explore ideas they might not have thought about otherwise,” says scientist Amy McGovern. As The Times adds, “A.I. hallucinations are reinvigorating the creative side of science. They speed the process by which scientists and inventors dream up new ideas and test them to see if reality concurs. It’s the scientific method — only supercharged. What once took years can now be done in days, hours and minutes. In some cases, the accelerated cycles of inquiry help scientists open new frontiers.”
M.I.T. Professor James J. Collins praises hallucinations for speeding his research into novel antibiotics. He notes that artificial intelligence models “come up with completely new molecules.” As The Times adds,
A.I. hallucinations arise when scientists teach generative computer models about a particular subject and then let the machines rework that information. The results can range from subtle and wrongheaded to surreal. At times, they lead to major discoveries.
In October, David Baker of the University of Washington shared the Nobel Prize in Chemistry for his pioneering research on proteins — the knotty molecules that empower life. The Nobel committee praised him for discovering how to rapidly build completely new kinds of proteins not found in nature, calling his feat “almost impossible.”
In an interview, Dr. Baker cited A.I. imaginings as key to “making proteins from scratch.” They have helped his lab obtain over 100 patents, mostly for medical care. One is for a new way to treat cancer. Another can be used against viral infections. Using his patents, Dr. Baker has helped start more than 20 biotech firms. “Things are moving fast,” he said. “Even scientists who do proteins for a living don’t know how far things have come.” His lab has designed “ten million” new proteins that “don’t occur in nature.”
More about how artificial intelligence is speeding up scientific discoveries is at this link.
