This new AI power race is inseparable from the positioning of AI as important for nationwide safety and US competitiveness with China. OpenAI laid out its place in a weblog put up in October, writing, “AI is a transformational know-how that can be utilized to strengthen democratic values or to undermine them. That’s why we consider democracies ought to proceed to take the lead in AI improvement.” Then in December, the corporate went a step additional and reversed its coverage in opposition to working with the army, asserting it might develop AI fashions with defense-tech firm Anduril to assist take down drones round army bases.
That very same month, Sam Altman stated throughout an interview with The Free Press that the Biden administration was “not that efficient” in shepherding AI: “The issues that I feel ought to have been the administration’s priorities, and I hope would be the subsequent administration’s priorities, are constructing out huge AI infrastructure within the US, having a provide chain within the US, issues like that.”
That characterization glosses over the CHIPS Act, a $52-billion stimulus to the home chips business that’s, a minimum of on paper, aligned with Altman’s imaginative and prescient. (It additionally preceded an government order Biden issued simply final week, to lease federal land to host the kind of gigawatt-scale knowledge facilities that Altman had been asking for.)
Deliberately or not, Altman’s posture aligned him with the rising camaraderie between President Trump and Silicon Valley. Mark Zuckerberg, Elon Musk, Jeff Bezos, and Sundar Pichai all sat straight behind Trump’s household on the inauguration on Monday, and Altman additionally attended. Lots of them had additionally made sizable donations to Trump’s inaugural fund, with Altman throwing in a $1 million private donation.
It’s straightforward to view the inauguration as proof that this cadre of tech leaders are aligned with one another, and with Trump’s orbit. However there are nonetheless some key dividing strains right here that can be price watching. Notably, there’s the conflict over H-1B visas, which permit many non-citizen AI researchers to work within the US. Musk and Vivek Ramaswamy (who’s, as of this week, not part of the Division of Authorities Effectivity) have been pushing for these visas to be expanded. This sparked backlash from some allies of the Trump administration, maybe most loudly Steve Bannon.
One other fault line is the battle between open- or closed-source AI. Google and OpenAI forestall anybody from realizing precisely what’s of their strongest fashions, usually arguing that this retains them from getting used improperly by dangerous actors. Musk has sued OpenAI and Microsoft over the problem, alleging that closed-source fashions are antithetical to OpenAI’s hybrid nonprofit construction. Meta, whose LLaMA mannequin is open-source, just lately sided with Musk in that lawsuit. Enterprise capitalist and Trump ally Marc Andreessen echoed these criticisms of OpenAI on X simply hours after the inauguration. (Andreessen has additionally stated that making AI fashions open-source “makes overbearing laws pointless.”)
Lastly, there are the battles of bias and free speech. As social media corporations have taken vastly completely different approaches to moderating content material—together with Meta’s latest announcement to finish its US reality checking program—it raises the query whether or not the best way AI fashions are moderated will proceed to splinter too. Musk has lamented what he calls the “wokeness” of many main AI fashions, and Andreessen stated on Tuesday that “Chinese language LLMs are a lot much less censored than American LLMs” (although that’s not fairly true, on condition that many Chinese language AI fashions have government-mandated censorship in place that forbids specific subjects). Altman has been extra equivocal: “No two individuals are ever going to agree that one system is completely unbiased,” he instructed The Free Press.
It’s solely the beginning of a brand new period in Washington, however the White Home has been busy. It has repealed many government orders signed by President Biden, together with the landmark order on AI that imposed guidelines for a way the federal government would use the know-how (whereas it seems to have saved Biden’s order on leasing land for extra knowledge facilities). Altman is busy as effectively. OpenAI, Oracle, and SoftBank reportedly plan to spend as much as $500 billion, on a three way partnership for brand new knowledge facilities; the brand new enterprise was introduced by President Trump, with Altman standing alongside. And, in response to Axios, Altman will even be a part of a closed-door briefing on January 30 with authorities officers, reportedly about OpenAI’s improvement of a robust new AI agent.
