OpenAI and Anthropic Lobby to Restrict Open-Source AI, Sparking Silicon Valley Backlash

OpenAI and Anthropic are reportedly lobbying Washington to restrict open-source AI models, sparking strong resistance from Silicon Valley and amplifying the AI safety debate around open-source regulation. Currently, US officials lean towards addressing Chinese open-source models as national security issues on a case-by-case basis rather than implementing a blanket ban.

Confirmed

According to the New York Times and sources close to the discussions, OpenAI and Anthropic have been lobbying US regulators to restrict open-source AI models. Officials are still weighing options but are more likely to treat Chinese open-source models as national security issues individually rather than imposing a comprehensive ban. Publicly, David Sacks defended the availability of open-source AI, arguing that US developers should retain the right to use public domain models, citing Cursor's Composer 2 built on Kimi K2.5. Additionally, AI safety and policy researchers, including Ryan Greenblatt, discussed open-weight models, affirming their value for safety research and understanding the current state of AI development.

Unconfirmed

Specific policy details regarding Anthropic's statements on Fox News and their push for restrictions remain subjects of debate. Furthermore, researchers (such as Nate) are divided on whether regulatory measures like download restrictions should be applied to open-weight models, with no consensus reached.

Why it matters

This controversy has evolved into a clear factional divide: on one side, Silicon Valley figures and developers advocating for the freedom to use Chinese open-source models (with Jensen Huang also publicly supporting open source); on the other, leading AI firms pushing for restrictions citing safety or national security concerns. This dispute extends beyond the right to use underlying code and models—it will directly impact the future trajectory of US AI open-source regulatory policy.

2026-07-23 ~ 2026-07-25 · 5 related posts

Primary sources

OpenAI and Anthropic reportedly lobby Washington to curb open-source AI models — firstadopter · 2026-07-25 23:04 (UTC+08:00)
David Sacks says open-source AI should stay usable for U.S. developers — beffjezos · 2026-07-25 15:04 (UTC+08:00)
AI Safety Researchers Debate Open Weights, Distillation, and National Security — dhadfieldmenell · 2026-07-25 04:23 (UTC+08:00)

Anthropic’s push for open-source restrictions is said to have united Silicon Valley against it — Hesamation · 2026-07-23
Anthropic gets dragged into a new open-source AI safety fight — mark_k · 2026-07-24
[source] AI Safety Researchers Debate Open Weights, Distillation, and National Security — dhadfieldmenell · 2026-07-25
[source] David Sacks says open-source AI should stay usable for U.S. developers — beffjezos · 2026-07-25
[source] OpenAI and Anthropic reportedly lobby Washington to curb open-source AI models — firstadopter · 2026-07-25