Join the Beta Waitlisteast

Explainable AI vs. Understandable AI

Knownwell

AI Knowhow Episode 61 Summary

  • Understandable AI and explainable AI are both important concepts in the field of AI
  • Explainable AI is a more technical term, whereas understandable AI is focused on helping end users know how/why an outcome has happened
  • Both are vital to developing trust with end users, which is a necessary foundation for AI uptake

Explainable AI vs. Understandable AI

What’s the difference between explainable AI and understandable AI? And why is comprehending the distinction between the two and when each is necessary important? That’s the topic of discussion in our roundtable discussion for this episode of AI Knowhow, with Knownwell CMO Courtney Baker, CEO David DeWolf, and Chief Product and Technology Officer Mohan Rao.

Mohan highlights that explainable AI primarily deals with the technical aspects of why a model behaves a certain way. “It’s like saying, ‘I did A and then I did B and then I did C to produce D,” Mohan says. “Understandable is always the goal, but sometimes it’s a lot harder because the context matters a lot.”

Understandable AI is geared toward creating intuitive AI systems that end users can easily comprehend without needing technical knowledge or deep explanations. One of the difficulties of ensuring understandable AI is that even developers and data scientists aren’t always sure what the outcomes of an LLM will be.

David underscores the importance of these concepts for business executives, emphasizing that trust in AI models will be a key hurdle to clear for them to achieve widespread adoption. Trust in AI systems is built through transparency, allowing users to know the inputs and processes that lead to an LLM’s decision while also fundamentally understanding its outputs.

He cites the Perplexity search engine as a real-world example of an AI platform that provides an experience that’s both understandable and explainable.

Practical Applications and Importance in Regulated Industries

Explainable AI is particularly important for companies deploying AI solutions in regulated sectors like healthcare and finance, where being able to document how an AI system reached its conclusion is critical. The primary aim of Explainable AI is to demystify complex AI models so that their decisions are more transparent, auditable, and trustworthy.

Both forms of AI are essential and interlinked, contributing to the development of trust and usability in AI-driven platforms, which in turn fosters broader AI adoption in businesses.

Expert Interview: Dom Nicastro on AI in Customer Experience

CMSWire Editor-in-Chief Dom Nicastro joins Chief Strategy Officer Pete Buer for an expert interview on the intersection of AI and customer experience. Dom shares valuable perspectives on the areas where he sees the greatest demand for AI to help drive improved customer experience.

The main use case where Dom is actively seeing demand for AI is for it to effectively be used to empower customer service agents by taking over mundane tasks, thus allowing agents to focus on delivering exceptional service. And in this day and age where the customer is king and choice is abundant, Dom says that the role of customer support has never been more important. “I think the support team is more important than the CEO in a company,” he says. To that end, it’s vital to equip support teams with the tools they need to be “specialized, skillful, thoughtful human beings” as Dom puts it.

From a journalism standpoint, Nicastro is optimistic. While AI can assist with generating content, the value of human insight and credibility remains irreplaceable. AI acts as an editorial assistant rather than a replacement, providing efficiency while maintaining the human touch necessary for credibility.

Watch the Episode

Watch the full episode below, and be sure to subscribe to our YouTube channel.

Listen to the Episode

You can tune in to the full episode via the Spotify embed below, and you can find AI Knowhow on Apple Podcasts and anywhere else you get your podcasts.

Show Notes & Related Links

You may also like