Then in late November, as a part of a broader effort to broaden and deepen public dialogue round science and AI, we co-hosted the AI for Science Discussion board with the Royal Society, which convened scientists, researchers, governmental leaders and executives to debate key subjects like cracking the protein construction prediction problem, mapping the human mind and saving lives by means of correct forecasting and recognizing wildfires. We hosted a Q&A with the 4 Nobel Laureates in attendance on the discussion board, Sir Paul Nurse, Jennifer Doudna, Demis Hassabis and John Jumper, which is obtainable to take heed to by way of the Google DeepMind podcast.
This was additionally a landmark yr for an additional cause: Demis Hassabis and John Jumper, together with David Baker, have been awarded the 2024 Nobel Prize® in Chemistry for his or her work on AlphaFold 2. Because the Nobel committee acknowledged, their work:
“[H]as opened up utterly new prospects to design proteins which have by no means been seen earlier than, and we now have entry to predicted constructions of all 200 million recognized proteins. These are actually nice achievements.”
It was additionally thrilling to see the 2024 Nobel Prize® in Physics awarded to not too long ago retired long-time Googler Geoffrey Hinton (together with John Hopfield), “for foundational discoveries and innovations that allow machine studying with synthetic neural networks.”
The Nobels adopted further recognitions for Google together with the NeurIPS 2024 Take a look at of Time Paper Awards for Sequence to Sequence Studying with Neural Networks and Generative Adversarial Nets, and the Beale—Orchard-Hays Prize, which was awarded to a collaborative crew of educators and Google professionals for groundbreaking work on Primal-Twin Linear Programming (PDLP). (PDLP, now a part of Google OR Instruments, helps clear up large-scale linear programming issues with real-world purposes from knowledge middle community visitors engineering to container transport optimization.)
AI for the good thing about humanity
This yr, we made quite a few product advances and revealed analysis that confirmed how AI can profit individuals instantly and instantly, starting from preventative and diagnostic drugs to catastrophe readiness and restoration to studying.
In healthcare, AI holds the promise of democratizing high quality of care in key areas, similar to early detection of heart problems. Our analysis demonstrated how utilizing a easy fingertip system that measures variations in blood move, mixed with primary metadata, can predict coronary heart well being dangers. We constructed on earlier AI-enabled diagnostic analysis for tuberculosis, demonstrating how AI fashions can be utilized for correct TB screenings in populations with excessive charges of TB and HIV. That is essential to decreasing the prevalence of TB (greater than 10 million individuals fall in poor health with it every year), as roughly 40% of individuals with TB go undiagnosed.
