In this episode of AI Equation, we dive into two compelling stories shaping the world of artificial intelligence. First, we explore President Biden’s groundbreaking executive order on AI regulations, which marks a significant leap in AI governance. The order focuses on safety, security, and reliability, mandating transparency for advanced AI systems, setting strict standards for comprehensive safety evaluations, addressing AI’s potential role in engineering dangerous biological materials, and combating AI-enabled fraud and deception. It also promotes equity, civil rights, and workforce impact mitigation while seeking international collaboration for trustworthy AI development.
Story 2 takes us into the ever-evolving landscape of AI, with a focus on Large Language Models (LLMs) and transformers’ rapid advancements. We delve into the dominance of GPT-4 and the ongoing debate surrounding governance and safety in AI. Openness and safety remain central themes, as we discuss the rise of open-source alternatives like Meta AI’s LLaMa model family. We also explore AI’s progress in various domains, from navigation and weather predictions to self-driving cars and music generation.
The episode highlights key takeaways from the report, including the ascent of compute as the new currency and the challenges in evaluating state-of-the-art models. GenAI startups, generative AI applications, and safety concerns in the AI community are also discussed.
Lastly, we feature Aidan Gomez, CEO and co-founder of Cohere, recognized as one of Time’s 100 most influential people in AI. Gomez co-authored the groundbreaking “Attention Is All You Need” paper, which introduced the transformer and revolutionized AI. He now leads Cohere, a company empowering businesses to integrate AI into their products. Cohere recently secured significant funding and aims to bridge the gap between AI theory and practical implementation.