AI gets its Nobel moment

This week, AI researchers celebrated significant achievements with the awarding of two Nobel Prizes, elevating their contributions to the forefront of scientific recognition.

The ongoing debate about AI's potential to transform the world takes a backseat to the immediate impact the technology is having in science, including solving complex problems and analyzing extensive scientific datasets. However, this progress also raises concerns about the misuse of advanced technology, Caliber.Az reports via foreign media.

While the groundwork for AI was laid over several decades, its recent advancements gained prominence, especially with the rise of chatbots and generative AI.

Geoffrey Hinton and John Hopfield received the Nobel Prize in Physics for their pioneering work on AI from the late 1970s to the 1980s. Both researchers utilized concepts from physics to develop artificial neural networks that have significantly shaped the AI landscape. Hinton is affiliated with the University of Toronto, while Hopfield is an emeritus professor at Princeton University.

In a related announcement, the Nobel Prize in Chemistry was awarded to Demis Hassabis (CEO of Google DeepMind), John Jumper, and David Baker for their groundbreaking work on protein structures essential to life. They were recognized for developing an AI system that solved one of biology's most challenging issues: predicting protein structures.

The Nobel Prizes are often given for research conducted long before, once its impacts are clearly established as beneficial to humanity. In a noteworthy move, the committee recognized the AlphaFold2 system, demonstrated only four years ago, which has aided scientists worldwide in various research endeavours. According to the committee, AlphaFold2 has been utilized by over two million people across 190 countries to address issues like antibiotic resistance and drug design.

Baker contributed to another AI-driven protein prediction tool, RoseTTAFold, and developed entirely new proteins. Hassabis has long been driven by a desire to create AI systems that facilitate scientific discovery.

Jumper noted that AI has advanced to the point where it aids scientists in designing better experiments through data insights. However, Hassabis emphasized that it remains premature to consider AI as a contender for all Nobel prizes, stating that human creativity and hypothesis development precede AI's role in data analysis.

Notably, three Nobel laureates have affiliations with Google. Hinton, who left the company last year, expressed concerns about the potential dangers of AI. These affiliations highlight the substantial resources required for AI research today, raising concerns about the concentration of power in profit-driven companies.

AI critic Gary Marcus pointed out that Hinton and others advocate for expanding AI via large neural networks, a foundation for generative AI. Conversely, Hassabis and his team are investigating neurosymbolic AI, which merges neural networks with symbolic knowledge. This approach gained attention in July when DeepMind announced its development of a math-capable AI system.

The ultimate effectiveness of either path remains uncertain, and there are no guarantees that either will prove advantageous for humanity.