Thomas Wolf is the cofounder and chief science officer of open-source AI platform Hugging Face, which provides access to thousands of pretrained AI models that can be downloaded and run locally. With over 10 million users, getting started on the site can be a daunting task. Thomas explains how the company aims to improve its accessibility through documentation on the company blog as well as community feedback, similar to social media likes and upvoting.

Thomas and Sam discuss the benefits and trade-offs of both open-source and closed-source AI models, as well as the evolution of microchips and the future of hardware and software development — as well as the hopes Thomas has for the future of coding with AI, starting with his children’s generation. Read the episode transcript here.

Guest bio:

Thomas Wolf is cofounder and chief science officer of Hugging Face, a collaborative AI platform. Wolf likes creating open-source software (OSS) that makes complex research, models, and data sets widely accessible. He can also be found pushing for open science in research in AI and machine learning, to try lowering the gap between academia and industrial labs through projects like the BigScience Workshop. He also writes and produces education content on AI, machine language, and natural language processing, including the reference book Natural Language Processing with Transformers, The Ultra-Scale Playbook, his blog, and videos.

Me, Myself, and AI is a podcast produced by MIT Sloan Management Review and hosted by Sam Ransbotham. It is engineered by David Lishansky and produced by Allison Ryder.

We encourage you to rate and review our show. Your comments may be used in Me, Myself, and AI materials.

Challenging the Average With Open-Source AI: Hugging Face’s Thomas Wolf