But Hodgkinson worries that researchers in the field will look to the technique rather than the science as they try to reverse-engineer why the trio won the award this year. “I hope this won’t cause researchers to misuse chatbots by mistakenly thinking that all AI tools are equivalent,” he says.
The fear that this will happen comes from the explosion of interest in other supposedly transformative technologies. “There are always hype cycles, most recently with blockchain and graphene,” Hodgkinson says. According to Google Scholar, after the discovery of graphene in 2004, 45,000 scientific articles mentioning the material were published between 2005 and 2009. However, after Andre Geim and Konstantin Novoselov received the Nobel Prize for the discovery of this material, the number of articles published then increased to 454,000 in 2010–2014 and over a million in 2015–2020. This surge in research has resulted in probably he only had humble real-world impact so far.
Hodgkinson believes that the energizing force of many researchers being recognized by the Nobel Prize panel for their work in artificial intelligence could cause others to rally around the field, which could result in a science of variable quality. “Do the proposals and conclusions have substance? [of AI] that’s a different matter,” he says.
We have already seen the impact of media and public attention on AI on the academic community. According to them, the number of publications on artificial intelligence tripled between 2010 and 2022 Stanford University researchwith almost a quarter of a million articles published in 2022 alone: over 660 new publications per day. This was before the release of ChatGPT in November 2022 ushered in the generative AI revolution.
The extent to which scientists are willing to follow media attention, money and praise from the Nobel Prize Committee is an issue that vexes Julian Togelius, a computer science professor at New York University’s Tandon School of Engineering who works on artificial intelligence. “Scientists generally follow some combination of the path of least resistance and greatest gain,” he says. Given the competitive nature of academia, where funding is increasingly limited and directly tied to researchers’ career prospects, it seems likely that it may be too tempting to combine a trendy topic that – as of this week – could win high achievers a Nobel Prize, to resist him.
There is a risk that this may impede new, innovative thinking. “Getting more fundamental data from nature and coming up with new theories that humans can understand are difficult tasks,” Togelius says. But this requires deep thought. It is much more productive for researchers to run AI-powered simulations that support existing theories and incorporate existing data, allowing for small advances in understanding rather than giant leaps. Togelius predicts that a new generation of scientists will eventually do exactly that because it’s easier.
There is also a risk that overconfident computer scientists who have helped develop artificial intelligence will begin to see that work on artificial intelligence is being rewarded with Nobel Prizes in unrelated fields of science – in this case, physics and chemistry – and decide to follow suit by stepping in into other people’s territory. “Computer scientists have a well-deserved reputation for sticking their noses into areas they know nothing about, injecting some algorithms and calling it progress, for better or for worse,” says Togelius, who admits he was already tempted to add deep learning to another field of science and “progress” in it before he considers it because he doesn’t know much about physics, biology or geology.