Open Source AI Goes Mainstream
Open source models are becoming increasingly mainstream as enterprises, governments, and startups alike embrace their promise of lower costs, greater control, and more flexibility. But as ever more capable models become freely available, a central question emerges: how do companies actually capture value? And with Chinese open models gaining ground, can American players keep pace?
Inside the Conversation
On this episode of TechCrunch's Equity podcast, Rebecca Bellan speaks with Irene Solaiman, Chief Policy Officer and SVP of Strategy at Hugging Face, about what startups need to know before building on open models—from choosing the right tools and infrastructure to navigating the trade-offs shaping the future of open AI.
Key Topics
- Who profits when the models are free? Exploring the business models emerging around open AI.
- Licensing and revenue sharing: How the Alibaba cloud advantage factors in.
- What founders should check first: Practical guidance before building on open models.
- The harness problem: Why the evaluation harness matters—and why no one can properly evaluate it.
- Buying GPUs vs. renting inference: The cost calculus facing startups.
- The Chinese model question: What happens if Chinese open models become the default?
- Beyond language: Nemotron, bio models, and the case for an American open frontier.
- Security pressures: The OpenCode attack and the push to close up, plus whether "obliterated" models serve defenders or attackers.
- Protecting small startups: Practical steps for staying secure.
Listen and Subscribe
Subscribe to Equity on YouTube, Apple Podcasts, Overcast, Spotify, and all the casts. You can also follow Equity on X and Threads at @EquityPod.
via TechCrunch AI