On diginomica, I've been pushing back on the generative AI hype machine with vigor, but also digging into its enterprise possibilities. Who better for an enterprise AI gut check than Vijay Vijayasankar? Vijayasankar has a deep interest/passion for AI but also has a reputation for debunking its misconceptions. In his role at IBM managing financial services clients, he can also speak to project realities - and how customers weigh GenAI risks and opportunities. I lay out my generative AI views/research for a gut check from Vijayasankar. We also discuss why he wrote his post on GenAI in the enterprise, and what he thinks customers should prioritize, as they move into GenAI projects. Vijayasankar also shares his own project learnings, and promising use cases. We hit on many issues in this podcast, such as whether bias and hallucations can ever be eliminated from LLM, and whether domain experts, customer data and reinforcement learning can improve the accuracy/effectiveness of GenAI for the enterprise. As Vijayasankar notes in the podcast, the views he expresses here, and in his post, are his views alone, not his employer. Vijay's AI post: https://andvijaysays.com/2023/08/05/genai-in-the-enterprise-nine-themes-that-i-have-seen-so-far/ My AI post: https://diginomica.com/enterprises-will-test-limits-llms-and-chatgpt-just-beginning

Ask this episode

“What did the council decide on item 42?” — answered with the speaker and the moment.

Transcript

Related