Summary
In this episode of Dr. Pawd, the host introduces Eric Horvitz, Microsoft's first Chief Scientific Officer, who has an extensive research background and is involved in various AI-related organizations and committees. The discussion delves into the National Security Commission on AI, established by Congress to focus on critical areas including AI research and development, trustworthiness, robustness, ethics, and engineering practices. The interdisciplinary commission's work has resulted in numerous legislative proposals and initiatives.
The conversation emphasizes the importance of AI researchers engaging with socio-technical issues and taking responsibility for their work's implications. The 100-Year Study on AI, a project dedicated to tracking AI's development and impact, is highlighted, with Stanford University committed to continuing the study. In conjunction, the AI Index, an annual project that delivers metrics and key takeaways related to AI research, industry, and applications, is also mentioned.
The podcast further discusses AI's centrality at Microsoft, its increasing importance across divisions, and the company's efforts in large-scale neural models, language models, and multimodal applications. Microsoft Research holds an annual meeting, the Disruptive Technology Review, where they examine key disruptive technologies with the potential to impact the company or the world. The growth of AI in the past decade is covered, with applications ranging from simple spam filters to complex memory management in operating systems.
Microsoft partners with OpenAI to be at the forefront of large-scale neural models, which require significant compute resources. The democratization of AI is a concern, and efforts like supporting university research and sharing AI resources with academic institutions help address this issue. Private-public partnerships and substantial public investments in AI play a crucial role in ensuring the continuous growth and development of the field.
The podcast highlights Microsoft's evolution under the leadership of CEOs Bill Gates, Steve Ballmer, and Satya Nadella, all sharing a passion for computing's potential to change the world. Microsoft's AI Ethics and Effects in Engineering and Research (AETHER) committee addresses safety, fairness, security, and privacy issues surrounding AI technologies. The sister organization, the Office of Responsible AI (ORA), supports these efforts by implementing responsible AI standards.
Lastly, the opportunities for AI in areas like advancing representation space, improving human-AI collaboration, incorporating causality reasoning, and its applications in biology, physics, and chemistry are discussed. This engaging conversation covers the ever-growing importance of AI in the world and the critical role researchers, organizations, and companies like Microsoft play in steering its development and impact on society.
