TLDR: The ‘AI2050’ Scholarship has announced its latest cohort of 28 scholars, who will receive over $18 million in funding to advance AI research across three critical areas: building AI scientist systems, developing secure and trustworthy AI models, and enhancing AI applications in biomedical research. The initiative, supported by Schmidt Science Fellows and funded by a former Google CEO, includes renowned scientists like Fei-Fei Li and Percy Liang as senior fellows, alongside promising early-career researchers and Tsinghua alumni.
The ‘AI2050’ Scholarship, a significant initiative aimed at fostering the inclusive and beneficial development of artificial intelligence, has unveiled its fourth cohort of 28 selected scholars. These researchers will collectively receive over $18 million in funding, distributed over three years, to pursue groundbreaking work in key AI domains.
The scholarship, backed by the Schmidt Science Fellows and funded by a former Google CEO, focuses on three pivotal research directions:
1. Building an AI Scientist System: This area aims to develop autonomous AI systems capable of collaborating with human researchers to accelerate scientific discovery.
2. Designing More Secure and Trustworthy AI Models: Research in this field is dedicated to creating AI systems that are inherently safer, more reliable, and transparent.
3. Enhancing the Application of AI in Biomedical Research: This direction explores how AI can be leveraged to drive innovations in healthcare and life sciences.
Among the 28 recipients, 21 are early-career researchers, and seven are distinguished senior researchers. The project has a global footprint, having funded 99 researchers across 42 institutions in eight countries since its inception in 2022.
Notable Chinese scientists, including Fei-Fei Li, often referred to as the ‘Godmother of AI,’ and Percy Liang, have been recognized as senior fellows. Chaowei Xiao, an alumnus of Tsinghua University, was also part of a previous ‘AI2050’ list earlier this year.
The selection process for senior fellows is based on their existing contributions through a closed nomination, while early-career fellows are required to hold postdoctoral or pre-tenure research positions. Beyond financial support, scholars gain access to annual academic exchange activities, fostering collaboration and knowledge sharing. The program also provides a special fund for significant computing needs, addressing hardware limitations to accelerate research.
Featured Senior Fellows and Their Projects:
Alán Aspuru-Guzik (University of Toronto): A professor of Chemistry and Computer Science, his research at the intersection of quantum information, machine learning, and chemistry aims to create an ‘AI Chemist’ (AIchemist). This AI scientist system, named ‘El Agente,’ will independently explore chemical spaces to discover new compounds, addressing global challenges like climate change and pandemics.
Surya Ganguli (Stanford University): A professor in Applied Physics and Deputy Director of the Stanford Human-Centered Artificial Intelligence Institute (HAI), Ganguli’s work spans AI, physics, and neuroscience. His AI2050 project seeks to establish a scientific foundation for interpretable and trustworthy AI by revealing the mechanisms behind generative models and large language models in creation, reasoning, and learning. The ultimate goal is to promote human-value-aligned AI systems.
Shirley Ho (Simons Foundation, New York University, Princeton University): A renowned astrophysicist and machine learning expert, Shirley Ho introduced 3D convolutional neural networks to astrophysics. Her AI2050 project focuses on scientific artificial general intelligence (scientific AGI), developing methods to understand and integrate various scientific AI models with language AI to create ‘world-understanding’ AI systems.
Sheila McIlraith (University of Toronto, Vector Institute): A professor of Computer Science and CIFAR Chair in AI, McIlraith’s research emphasizes human-centered AI and sequential decision-making. Her AI2050 project aims to endow AI with ‘Purposeful Theory of Mind,’ enabling AI to understand and consider the mental states, beliefs, desires, and intentions of itself and others, fostering trustworthy and cooperative intelligent agents.
Dawn Song (University of California, Berkeley): A professor of Computer Science specializing in AI security and assurance, Song’s AI2050 project is dedicated to developing ‘secure and verifiable’ AI tools. These tools will automatically generate code along with formal security specifications and mathematical proofs, fundamentally eliminating security vulnerabilities and enhancing the trustworthiness of AI systems for critical applications.
Also Read:
- UNESCO Inaugurates Beruniy Prize to Advance Ethical AI Innovation and Global Cooperation
- Aston University Launches Prestigious AI Fellowships for Postdoctoral Researchers
Philip Torr (University of Oxford): A professor with extensive experience in computer vision and machine learning, Torr’s research at the Torr Vision Group (TVG) has made significant contributions to computer vision.
