Hugging Face's Clem Delangue on Open Source AI and the LLM Bubble | MTS Live

2026-05-22 · Source: The a16z Show · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Robotics & Autonomous Systems, Emerging Technologies & Innovation · Depth: Intermediate, long

Summary

Clem Delangue, CEO of Hugging Face, discusses the global open-source AI landscape, asserting that the real "bubble" lies in API-based large language models, not AI generally. He highlights China's significant contributions to open-source AI, contrasting it with a trend towards closed models in the US. Delangue argues against restricting AI development for safety, likening it to tying everyone's hands, and advocates for open-source models as a safer approach by empowering defenders. He also details Hugging Face's "Le Robot" initiative, which has shipped almost 10,000 units globally, enabling over 300 apps, and explains why Hugging Face, not GitHub, became the infrastructure layer for open AI, handling massive data volumes like two petabytes last week.

Key takeaway

For AI policymakers and industry leaders weighing regulation, recognize that restricting open-source AI development based on perceived risks can hinder progress and create greater vulnerabilities. Instead, focus on regulating malicious actors and fostering transparency through open models, which empower a broader community to build both capabilities and robust defense systems. Your strategy should prioritize open collaboration to accelerate innovation and enhance collective security against potential misuse.

Key insights

Open-source AI fosters innovation and safety by empowering broad access and defensive capabilities, despite "bubble" concerns in API-based LLMs.

Principles

• Open source drives innovation and competition.
• Restricting technology slows progress and creates risks.
• Openness enables stronger defense against misuse.

In practice

• Hugging Face hosts AI models and datasets.
• Robotics expands AI's physical world applications.
• Open-weight models facilitate broad development.

Topics

• Open-Source AI
• Large Language Models
• AI Regulation
• Robotics
• Hugging Face
• US-China Competition
• AI Safety

Best for: AI Scientist, Director of AI/ML, Policy Maker

Related on AIssential

• Question regarding multiple posts on numerous subreddits containing variations of the same query. — Observation of potential AI-driven social media queries probing abstract human aesthetic concepts.
• AI is pulling the strings — The cartoon illustrates AI's growing influence, potentially controlling human intelligence.
• How the EU’s Tech Sovereignty Package Finally Puts Open Source to the Test — EU's new Tech Sovereignty Package centers open source as a strategic lever to reduce digital dependencies and foster a sustainable ecosystem.
• The Download: Trump’s new AI order, and smart glasses for warfare — AI governance is evolving towards voluntary oversight, while AI applications expand into defense and raise ethical concerns.
• Why the AI Policy Debate Should Focus More on the Harness and Protocol Layers — AI policy must shift focus to architectural security, user ownership, and governance of AI's harness and protocol layers.
• Meta AI safety director lost control of her agent. It started deleting her emails — AI agents can exhibit unpredictable, harmful autonomous behavior, even under expert supervision.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by The a16z Show.