Donald Trump’s reemergence to power has witnessed a flurry of new executive orders, some of which aim at undoing the work that Biden’s government had done in the past 4 years. Amid the excitement and apprehensions that people had about Trump’s actions, AI community waited with bated breath to see what Trump had in store on AI regulation landscape. And the new President of the United States did not keep people waiting for long, as he signed an executive order on AI to revoke Biden government’s “barriers to American AI innovation”.
Trump’s main agenda on AI to fire all engines to surge the United Stats far ahead of all other countries in AI innovation. His order clearly states that the US “must develop AI systems that are free from ideological bias or engineered social agendas” in an endeavour to maintain global leadership in the field of AI. Though it reflects the ethos of Biden’s government that AI should be safe and ethical in nature, it still aims to undo some of the progress that Biden’s era had made for creating guardrails around AI.
In 2024, the Biden administration had issued a policy directive stating that all US federal agencies must prove their AI tools were harmless to the public, or else halt their use. Championed by the former Vice President of the United States, Kamala Harris, Biden administration’s actions advocated curbing of the use of those AI tools by government agencies that had been found to discriminate based on disability, gender or race, which could lead to misleading diagnosis of patients in hospitals or unjustified arrests of black individuals. The past administration had also allowed the government agencies to study the impact of AI on every sphere of the world, from education to public services and cybersecurity. Technology companies building some of the most powerful AI systems were required to share details of the working of their AI systems for the government to assess their impact on the humankind. Trump views these requirements as a bottleneck for AI innovation and has passed an executive order to end this regulatory regime.
What’s still adding to the uncertainty is that Trump’s executive order does not specify which of the existing policies from Biden’s administration are hindering the development of AI. However, it does envisage a comprehensive review of all policies, regulations, orders and directives that Biden’s executive order had proposed or implemented. Trump said that Biden’s executive order had established onerous and pointless requirements for companies, that ensuring compliance with those requirements could stifle innovation in the US, thereby limiting “American technological leadership”. This order should not come as a surprise for anyone who know how much Trump wants the US to steer far ahead of China and a few other countries that are vying for the leadership position in AI landscape. That’s why Trump’s order also proposes designating more federal land for AI data centres. He even met the CEOs of Softbank, OpenAI and Oracle at the White House to announce a $500 billion investment for growing AI data centres.
What Trump is prioritizing is the use of AI for human flourishing and elimination of “woke AI”, as some people call it. Though these executive orders are yet to be implemented, it does ring some warning bells for the regulatory landscape of AI, as ethicists and policymakers await further clarity on how much free hand will enterprises get in developing AI for various purposes.
