15 November 2024
The results of the US House of Representatives election were called yesterday, showing that the Republican Party will control the House. This means that President Trump’s party will control the Senate, House, and retain its majority of Supreme Court Justices.
President Trump has also confirmed the names of several proposed cabinet members, giving us greater insight into his policy priorities.
Proposed appointments include Secretary of State Marco Rubio (Florida senator), Defense Secretary Pete Hegseth (Fox News host), Attorney-General Matt Gaetz (former Florida congressman), Department of Government Efficiency Elon Musk and Vivek Ramaswamy, Director of National Intelligence Tulsi Gabbard (former Democratic congresswoman), Health and Human Services Secretary Robert F Kennedy Jr, and White House Chief of Staff Susie Wiles, with Stephen Miller as Deputy.
What does this mean for the future of AI and technology regulation in the US and rest of the world?
AI safety and Executive Order 14110
President Biden’s Executive Order on AI Safety was a landmark in US AI regulation, leading to major policy developments including requiring foundation model developers to inform the Department of Commerce before providing models to foreign customers.
According to Time Magazine, “Trump promised to repeal the Executive Order on the campaign trail in December 2023, and this position was reaffirmed in the Republican Party platform in July, which criticized the executive order for hindering innovation and imposing “radical leftwing ideas” on the technology’s development.”
However, with Elon Musk repeatedly arguing for greater AI safety, including supporting the Calfornia AI Safety Bill that Governor Newsom vetoed in September, some commentators believe that Trump could pass an executive order going even further than Biden’s on national security.
Samuel Hammond, a senior economist at the Foundation for American Innovation told Time Magazine that: “It would not surprise me if a Trump executive order on AI maintained or even expanded on some of the core national security provisions within the Biden Executive Order, building on what the Department of Homeland Security has done for evaluating cybersecurity, biological, and radiological risks associated with AI”.
The US AI Safety Institute (AISI) created by the Biden Administration in November 2023 has attracted bi-partisan support from Congress, with members of the Senate and House supporting calls from industry to place the AISI on a legislative footing, according to Hamza Chaudhry, U.S. policy specialist at the nonprofit Future of Life Institute.
Perhaps a telling sign of the potential move towards greater AI safety is OpenAI CEO Sam Altman’s recent comments on Artificial General Intelligence. In a seeming reversal from his previous views on the potential of AGI to transform the world, Altman recently told Reddit that AGI will be achieved in the next few years but will bring “surprisingly little” societal change. Coming after Trump’s election with Musk in full support, perhaps the downplay of AGI’s impact may be designed to reduce pressure from a pro-safety administration…
Platform regulation and online safety
Trump’s close relationship with Elon Musk, who reportedly event sat in on calls to Heads of State, could indicate that US attempts to regulate social media (such as the Kids Online Safety Act) will stall, signalling even further divergence from EU and global attempts to regulate online platforms, such as the EU Digital Services Act. However, there are indications that Musk’s relationship with Trump is already souring, with former White House Press Secretary Anthony Scaramucci in particular predicting that a rift will soon occur.
Speaking to friends who are strong Trump supporters (yes, I do have friends who support Trump…ending polarisation means listening to others’ views!), one of their key concerns is government attempts to control social media.
The Supreme Court’s recent ruling on state power to regulate online platforms, Moody v NetChoice LLC, which held that the First Amendment takes priority over laws requiring platforms to remove particular content or ban users, may be an indication of the likely policy direction of the Trump Administration.
According to the Moody judgment: “…this Court has many times held, in many contexts, that it is no job for government to decide what counts as the right balance of private expression—to “un-bias” what it thinks biased, rather than to leave such judgments to speakers and their audiences. That principle works for social-media platforms as it does for others” (at page 4).
We can expect greater moves to de-regulate online content from the next administration, although the influence of private companies' algorithms may still come under scrutiny if conservative views are seen to be sidelined.
Government 'waste' and budget cuts
Elon Musk and Vivek Ramaswamy have been selected to head the new Department of Government Efficiency (‘Doge,’ also the name of a cryptocurrency supported by Musk). The ‘department’ will not be an official government agency – which would require legislation – but will instead “provide advice and guidance from outside of government” according to Trump. Musk has pledged to reduce federal government spending by “at least $1 trillion,” almost a third of the total US federal government budget.
Similarly, Trump’s selection for Attorney General Matt Gaetz led the push to reduce FBI funding by $375 million last year when he was a Member of Congress (a position he has resigned to take on his new role for the Trump Administration).
Since the EU-US Data Protection Framework depends upon a team of lawyers and policy staff at the Department of Justice and in the intelligence agencies, any major cuts to these departments could threaten the data transfer mechanisms established following the Schrems II decision.
With the Federal Trade Commission and SEC acting as de facto privacy regulators in the US, any major cuts to these departments could signal a reduction in pro-privacy investigations at the federal level.
Antitrust and big tech
Trump will also be able to select a new chair of the Securities and Exchange Commission (SEC) and exercise control over the Justice Department and FTC, both of which are currently involved in major antitrust cases such as that against Google.
According to Reuters: “The FTC and DOJ would likely scrap merger review guidelines crafted under Biden, said Jon Dubrow, a partner at law firm McDermott Will & Emery.”
Reuters also reports that the final stages of the Google antitrust case, due to take place in April, are now less likely to result in the company’s break-up, following comments made by Trump at rallies in October.
Trade and tariffs
Trump has selected Robert Lighthizer as trade envoy: a proponent of “America First” agenda, with proposals to impose tariffs of up to 10-20% on imported goods from most countries (including the EU) and up to 60% on China.
Tariffs on EU and UK goods could be as high as 10-20%, potentially resulting in reduced trade between the EU and US. If the EU is seen as a less important market for the US, could this mean that the AI Act (which will impact AI system providers based in the US) will not achieve the ‘Brussels effect’ EU legislators are hoping for?
Tariffs on the EU and UK could also see far stronger trade links between the UK and the bloc, perhaps resulting in closer regulatory alignment; this could be crucial as the UK government considers how to regulate AI including legislative options.
Trade and microchips
The Biden Administration CHIPS Act, which introduced guardrails to restrict exports of superconductors to China, was met with China’s restrictions on exports to the US preventing export of certain microchips to US companies.
Recently, Taiwan’s TSMC halted exports of its advanced chips to Chinese companies in response to US government demands, after their chips were found in Huawei products.
Superprocessors and superconductors are the bedrock of all AI and advanced computing technology. A trade war between the US and China in this market could result in serious restrictions in technological advancements.
However, whilst it is predicted that Trump would continue to push for restrictions on China, he also commented on the Joe Rogan podcast in October that Taiwan should begin to pay the US for its defence and that it is “stealing” the US microchip market.
The potential for greater US protectionism may be strong, but the US reliance on, and support for, Taiwan (which still produces 96% of the world’s superprocessors) remains unclear.
Trade and autonomous vehicles
On autonomous vehicles, Trump’s trade policies seem clearer, threatening to impose “100, 200, 2,000%” import tariffs on Mexican and Chinese vehicles, which could have a major impact on manufacturing in the US and beyond.
Depending on the details of the tariffs, imposing such huge restrictions on importing vehicles is likely to have a major impact on the autonomous and electric vehicle industries in the US, and signals a potential rift in the development of international standards on autonomous vehicles.
Put simply, if the US no longer imports Chinese cars, there will be little appetite to develop interoperable standards.
AI and warfare
Trump’s pick for Defense Secretary Pete Hegseth has suggested that US soldiers should be pardoned for war crimes, possibly indicating a more hawkish stance on the use of advanced warfare strategies such as autonomous weapons.
On 4th November, the New York Times reported that Meta was allowing the Pentagon to use its AI models for defence purposes, perhaps an indication of the closer alignment between the military and big tech.
Elon Musk’s Starlink is a significant supplier of satellite technology for the Pentagon and US military; will cuts to federal spending be linked to an increase in use of the private sector for military technology?
AI and immigration
Proposed Border Tsar Tom Homan has pledged to remove up to 20 million people living in the US as undocumented immigrants, and increase border checks for those entering the country.
The scale of the task in identifying and removing such a large number of people is likely to involve greater use of immigration technology, particularly facial recognition and biometric checks.
Use of AI for facial recognition and ‘real-time’ biometric identification is increasingly restricted in the EU. Article 5 of the AI Act, which lists prohibited uses of AI systems, includes prohibitions on “the use of AI systems that create or expand facial recognition databases through the untargeted scraping of facial images from the internet or CCTV footage” (see also Recital 43) and certain “‘real-time’ remote biometric identification systems in publicly accessible spaces for the purposes of law enforcement, unless and in so far as such use is strictly necessary” for specified purposes (see Recitals 32, 33, 38, 39, 40 and 41).
Could the use of AI for immigration and border controls see a major divergence in not just the regulation of AI, but also the development of AI technology in the US and EU?
These are just a handful of the policy areas impacting AI and data protection that could see significant changes in the next four years.
Although nothing is certain at present, there are clear underlying tensions in Trump’s policies: in particular the drive to reduce the role of the state on the one hand, and the push to protect free speech and US economic interests on the other.
However, as the new administration takes office, it may be that these predictions fall away as plans come into contact with reality. To paraphrase Sam Altman’s recent comments on AGI, maybe the second Trump Administration will come and go in the next few years with “surprisingly little” societal impact...