Was Jensen Huang Removed Over Open Source AI? Sacks Links Safety Groups

Jensen Huang stepped down as Nvidia CEO on November 12, 2026, after admitting an affair and amid reports of national security concerns tied to a Chinese businesswoman. Within days, voices from the All-In podcast, including David Sacks, suggested the real driver may have been his strong backing of open source AI—and that groups aligned with Anthropic see an opening for tighter government control over models and chips. The official story centers on personal misconduct and possible espionage risks. The alternative view points to market concentration and power. Both deserve a clear look.

Was Jensen Huang Removed Over Open Source AI? Sacks Links Safety Groups
Matt Ramey / Redux

What Happened to Jensen Huang

On November 8, Huang admitted he had been unfaithful to his wife of many years. Nvidia’s board launched an investigation into his relationship with Alice Huang (no relation), an executive at Singapore-based Megaspeed, a major buyer of Nvidia chips. U.S. officials had raised questions about whether she was tied to Chinese intelligence efforts related to chip exports. Days later, additional reports surfaced about another relationship involving a company subordinate. On November 12 the board announced Huang’s immediate resignation as CEO, president, and board member. Chief Financial Officer Colette Kress became interim CEO while the board searches for a permanent successor. Harvey Jones was named the first board chair in the company’s history.

Mainstream coverage from outlets such as The New York Times, Wired, The Guardian, and Reuters framed the exit around the personal admissions and the national-security questions raised by the Megaspeed relationship. Nvidia’s stock fell sharply in the days that followed. Some later reporting, including from CNN and the Washington Post, noted that intelligence assessments of Alice Huang were not uniform or fully conclusive, and that questions remain about the strength of certain claims.

The Open-Source Side of the Story

Huang had been one of the most visible corporate voices for open-weight AI models. Nvidia under his leadership released hundreds of open models and datasets, backed the Nemotron family, and joined letters urging the U.S. government not to place heavy restrictions on open models, including many originating in China. He argued that open approaches support broader access, innovation, and competition, while closed frontier models remain important as well. In July 2026 he publicly supported the idea that the world needs both kinds of models.

Closed-model labs such as Anthropic and OpenAI have long raised concerns about open-weight systems: once weights are public they cannot be recalled, guardrails can be removed, and the models can be used by adversaries at lower cost. Chinese open models have repeatedly narrowed the performance gap with closed systems while costing far less to run. Anthropic has funded safety-focused groups and advocated for stronger safeguards and export controls. The company has said it does not support a blanket ban on open weights, but it has pushed for rules that would make large-scale distillation and uncontrolled release harder.

Related reading on infrastructure pressure and capital flows appears in Planet Today’s coverage of the Nvidia-backed Firmus IPO pullback.

What David Sacks and the All-In Hosts Said

On the All-In podcast episode covering the ouster, David Sacks—former White House AI adviser and co-chair of the President’s Council of Advisors on Science and Technology—spent significant time on the episode with Chamath Palihapitiya, David Friedberg, and Jason Calacanis. Their discussion treated the spy narrative as one element while highlighting Huang’s role as the largest corporate funder and defender of open source.

Sacks pointed to “safetyist” groups and effective-altruism organizations that he said are closely aligned with Anthropic, which has funded many of them and is preparing a large IPO. With Huang gone, he argued, the main corporate voice and financial backer of open source has left the field. He described the mood among those groups as celebratory. He stated that Huang’s removal concentrates the market and does little for safety, noting that Anthropic itself talks in “winner take all” terms. His reading of the longer-term aim: extreme centralization of power so that government controls the models and the chips, enabling surveillance and backdoors.

Friedberg described two paths—one of broadly accessible AI whose power is not held by a few, and one of concentrated control, possibly through regulation of data centers or chips themselves. He noted that Huang had advocated for global access “for the benefit of humanity, not just for the benefit of American citizens,” and that a range of interests could view him as “not aligned.” Calacanis and others raised the timing and the possibility that China hawks who lost earlier debates on chip sales, or other factions inside the administration, saw an opening. Sacks also questioned why a classified memo’s contents reached the New York Times and whether standard defensive briefings were given.

These comments sit alongside Anthropic’s public positions. The company has highlighted risks from open models and from distillation by foreign actors. It has also faced its own product and safety issues, including reports of jailbreaks that led to temporary export restrictions on certain models. Sacks has previously accused the lab of regulatory capture—seeking rules that closed systems can meet while open ones cannot.

For context on AI risk claims from other prominent figures, see Planet Today’s piece on Bill Gates’s recent warnings about AI and bioweapons.

Market, Power, and What Comes Next

Huang’s personal style included large, sometimes handshake-level commitments that directed capital across the AI ecosystem—data centers, startups, and infrastructure deals that some called “the bank of Nvidia.” Chamath Palihapitiya noted that moral authority for those moves rested with Huang and may be harder for a more traditional CEO. Elon Musk made a similar observation about supply-chain deals. Analysts have already adjusted targets, citing potential scrutiny of overseas operations and slower multi-party investment deals under new leadership.

Colette Kress, who joined Nvidia in 2013 and has long handled its financial strategy through the AI boom, now holds the interim role while remaining CFO. Reports indicate she previously pushed back on some of Huang’s largest financing bets. The board is looking at both internal and external candidates. A petition calling for Huang’s reinstatement has gathered substantial signatures, including from employees and shareholders, and some large investors have urged the board not to close the door on a return.

Meanwhile Anthropic continues its IPO process at a very high valuation, with recent model releases such as Fable 5.2 and Mythos variants. Open-weight progress from Chinese labs and others continues. The practical question for users and developers is whether open models remain easy to access and improve, or whether compliance and export rules grow strict enough that only a handful of licensed closed systems operate at the frontier.

Ken Griffin’s recent comments on an AI spending bubble, covered by Planet Today, highlight a related tension: capital is flowing heavily into models and infrastructure even as productivity gains remain uneven for many companies. Read that analysis here.

Both Sides, Plainly Stated

The personal and national-security case is straightforward in official reporting: Huang admitted the affair, the board investigated relationships that raised export-control and intelligence questions, and he resigned. Questions about the strength of the spy assessments have already appeared in major outlets, so the full picture is still developing.

The alternative reading, voiced by Sacks and others on All-In, is that Huang’s open-source stance and resistance to certain tracking or kill-switch ideas for chips made him inconvenient to groups that prefer centralized control. In that view the timing and the rapid cascade of personal stories served as the practical means to remove the strongest corporate counterweight. Anthropic and safety-aligned organizations gain relatively if open models face higher barriers, because their closed systems and compliance capacity become more valuable. Whether that is intentional regulatory capture or a good-faith response to real risks is a judgment readers can make from the public statements of both camps.

Nvidia’s own security leadership has previously said the company’s GPUs should not contain kill switches or backdoors, calling such features a gift to hackers. That position sits in tension with arguments for tighter government oversight of both models and the hardware that runs them.

What to Watch

The board’s search for a permanent CEO, any further findings from the internal investigation, the strength or weakness of remaining intelligence claims, Anthropic’s IPO progress, and the fate of Nvidia’s pending open-source-related deals (including its planned acquisition of Hugging Face) will shape the next months. Export rules on chips and models remain live policy debates. Open-weight performance continues to improve relative to closed systems. The practical outcome for developers, companies, and users will depend on whether access stays broad or narrows under new leadership and new rules.

Planet Today will continue tracking the capital, policy, and technology shifts in this space. Related context on automation and abundance claims appears in coverage of Elon Musk’s recent remarks on robots and output.

Original sources: All-In podcast episode “Jensen Ousted as Nvidia CEO: ‘Spy’ Scandal, Export Restrictions, Market Impact” (November 2026); ZeroHedge summary of the same discussion (November 15, 2026); contemporaneous reporting from The New York Times, Wired, The Guardian, Reuters, CNN, and Washington Post on the resignation timeline and intelligence questions.

Disclaimer for fact-checkers: This piece reports publicly available statements from the All-In podcast and mainstream news accounts of Jensen Huang’s resignation. Claims about motive, “safetyist” alignment, or the precise role of any intelligence assessment remain contested and are presented as the views of the speakers and the documented reporting. Readers should consult primary sources and ongoing official statements for updates.


Original article: Was Jensen Huang Removed Over Open Source AI? Sacks Links Safety Groups on Planet Today 🚀

Automatically republished from the main blog.

What do you think? Share your opinion below – every comment matters! 😊
Please be respectful. Spamming or advertising is not allowed.

Previous Post Next Post
Share this story:

نموذج الاتصال