On November 12, 2026, Erik Voorhees, founder of the open‑source AI service Venice, warned that giving politicians a veto over frontier artificial‑intelligence models would concentrate power in a single governmental body. His comments followed a recent All‑In podcast clip in which co‑host David Sacks said Democrats are likely to push for a federal department that would review and potentially block the release of advanced AI systems.

Legislative proposals targeting AI releases

The warning references two related legislative efforts. In September, Senator Bernie Sanders and Representative Greg Casar introduced the Ban Artificial Superintelligence Act (S.5493). The bill would establish a Department of Artificial Intelligence, require developers to obtain approval before deploying, importing, or transferring an “advanced AI system,” and give the secretary of the department authority to pause development if the model exhibits dangerous characteristics such as superintelligence potential. The proposal also includes Senate confirmation of the secretary, independent scientific advice, and a limited right of appeal to the Federal Circuit.

Separately, Abdul El‑Sayed’s October 13 campaign platform, titled “AI Under Control,” proposes a Cabinet‑level agency with similar inspection and block powers. The platform builds on the Sanders‑Casar bill, adding mandatory pre‑release audits, 24‑hour breach reporting, and whistle‑blower protections. While both proposals grant a government veto, they differ on governance structures and the role of external scientific boards.

Voorhees’ argument for open‑source decentralization

Voorhees, who launched Venice in July 2024 to provide open‑source AI tools, argues that licensing and centralized approval mechanisms undermine equality and transparency. He likens the concentration of AI gatekeeping power to a fictional “Sauron” holding a ring, emphasizing that any single entity—whether a government agency or a large closed‑model company—could dictate the direction of AI development. In his view, decentralized, open‑source models offer a safer path because they allow broader scrutiny and community‑driven safety measures.

Political context and upcoming test

The debate occurs against a shifting political landscape. According to the Associated Press, Democrats secured control of the House and unified Congress in the November 2024 elections, raising the likelihood of a coordinated AI regulatory push. However, the proposals remain bills and platform statements; they have not yet been enacted into law. President Trump’s June 2, 2024 executive order on AI continues to promote a voluntary framework rather than mandatory licensing, highlighting the current policy split between voluntary industry standards and proposed compulsory oversight.