#246: Unlocking Value with Generative AI

In this episode, Kendra Vant and Tracy Moore delve into the world of generative AI and its potential for unlocking commercial value. They kick off by addressing the excitement and hype surrounding generative AI technologies and emphasise the importance of grasping the fundamentals to extract real value from these advancements. 

Next, the discussion moves to AI as a general-purpose technology, where different perspectives on the speed and extent of its transformative potential are explored. Realistic expectations are highlighted, with comparisons drawn to the gradual adoption of internet banking. Felipe, Kendra and Tracey also discuss the shift from perceiving technology as "magic" to a practical "tool."

The probabilistic nature of generative AI and its implications for user interactions are explored; Managing expectations and redesigning products to align with this probabilistic nature is deemed essential.

Kendra and Tracy share insights into effectively leveraging large language models (LLMs) for product development. They stress the significance of asking the right questions, defining clear objectives, and understanding failure modes. Challenges related to managing costs, data requirements, and GPU resource availability are also discussed.

Navigating different phases of product development, from ideation to execution, is a complex task. Kendra and Tracy emphasise experimentation, continuous evaluation, agility, communication, and cross-functional collaboration as key factors for success.

Effective communication within interdisciplinary teams working with AI is paramount. They discuss the importance of bridging knowledge gaps and overcoming vocabulary barriers to ensure that everyone involved understands the goals and processes.

Their valuable insights shed light on the potential of this technology for commercial applications.

Enjoy the show!

 

Thank you to our sponsor,⁠ Talent Insights Group⁠!

Join us for our next event: OpsWorld: Deploying Data & ML Products (Melbourne, 24-25 October): ⁠https://www.datafuturology.com/events⁠ 

Join our Slack Community:⁠ https://join.slack.com/t/datafuturologycircle/shared_invite/zt-z19cq4eq-ET6O49o2uySgvQWjM6a5ng⁠ 

 

“If there's one piece of advice I could give to anyone who's starting down the path of building AI into their products, it's have a conversation with all of the folk involved in your team about this will be successful if we can all understand all the pieces that we're bringing to the puzzle.”

- Kendra Vant, Author of Data Runs Deep, ex Xero, ex SEEK, advisor to the National AI Centre & National Data Commissioner

 

WHAT WE DISCUSSED

00:00 Introduction of Kendra Vant

4:21 Introduction of Tracy Moore

7:44 Discussion on the potential and natural evolution of generative AI technologies.

8:21 Exploration of the current state and potential future of generative AI technologies.

15:39 Discussion on the limitations and challenges of integrating generative AI technologies into workflows.

20:06 Discussion on the suitability of generative AI technologies for specific product experiences.

21:26 Exploration of the challenges and considerations in building successful systems using generative AI technologies.

26:31 Importance of clear communication and dialogue in defining goals and integrating generative AI technologies into user experiences.

30:59 Importance of experimentation, measurement, and promotion of successful generative AI implementations.

34:24 Discussion on the need for governance, customer feedback, and adapting organisational structures for generative AI technologies.


At Data Futurology, we are always working to bring you use cases, new approaches and everything related to the most relevant topics in data science to help you get the most value out of these technologies! Check out our upcoming events for more amazing content. And as always, we appreciate your Reviews, Follows, Likes, Shares and Ratings. It really helps new data scientists find us.