Donald Trump’s government hasn’t been shy about embracing generative AI. No later than his second day in office, the President announced the $500 billion Stargate project: a joint private venture aimed at constructing 20 AI datacenters in the US. Standing beside Trump when he made the announcement were the leaders of OpenAI, Oracle, and Softbank — three AI-focused companies backing the project. These leaders have made it clear that they stand behind Trump: literally, financially, and morally. Where he goes, they go too.
That’s unimaginably dangerous.
Generative AI is a powerful tool, but it is a tool that can be used for good or ill. Currently, the US government wields that tool. Instead of promoting the general welfare, the government is now acting against its own people, and AI is helping it to do so — primarily through surveillance.
The government is already using AI as a surveillance tool within the government itself. Elon Musk’s Department of Government Efficiency weaponizes AI to spy on the communications of employees of the EPA, looking for any anti-Trump or anti-Musk sentiment. The resemblance to Orwell’s 1984 is uncanny: a tyrannical, technology-wielding government suppressing its employees. This sort of government surveillance is something that one hopes remains fiction — but it has leapt out of the pages of dystopia and into the real world. It’s only a matter of time before such measures will be employed against the general populace, rather than being confined to federal workers. Indeed, it may already be happening.
Multiple government agencies monitor social media — most recently, US Citizenship and Immigration Services announced that it would begin monitoring the social media of immigrants. And how does it accomplish this monumental amount of surveillance? Likely through the usage of AI. The US has at least 4 contracts with social media monitoring companies — and all of these companies use AI as a part of their toolkit. AI, in this context, is more than just a search tool or a means of organization. Rather, it stands in for what a human should typically be doing: critical thinking skills and judgement calls.
Allowing AI to perform these inherently human functions had already led to false arrests as a result of something called “automation bias.” This refers to the human tendency to trust technology to deliver correct information — an issue that resulted in the mistaken arrests of at least eight people in the US. Even worse, AI surveillance intensifies racial profiling — as in nearly every one of these cases, the person arrested was Black. Law enforcement relying on AI poses catastrophic risks and empowers AI companies to decide who goes to jail. Certainly, they are already deciding whose voice gets silenced, specifically in the world of social media.
To no one’s surprise, social media is a major culprit for political polarization in the US today. For the past decade, misinformation — particularly relating to the 2020 election — has become commonplace on social media outlets. Coupling that with Elon Musk’s usage of X (formerly Twitter) to promote his own right-wing views, it’s clear that social media is quickly becoming a means to disseminate propaganda. The engines driving this rapid proliferation of right-wing content are the algorithms of these respective sites, which determine which person sees what content, and when. Those algorithms are, like so much else nowadays, powered by AI.
The danger here is glaring. It seems logical that a company aligned with America’s right wing would have everything to gain by filtering right wing content into the feeds of millions of people. And at this most critical juncture, there is almost nothing in place to stop them from doing so. Though there’s no way to be certain as to whether this is already happening, one can already see potential signs of AI being used to promote right-wing viewpoints, particularly by the US government itself.
The official White House X account has posted art created by AI at least three times now — with one post depicting Trump with a crown on his head and another showing an immigrant being arrested. These images are chilling indicators of the Trump government’s vision for AI and a glimpse into the minds of those responsible for creating such images. The first image, that of Trump styled as a “king,” promotes authoritarianism, plain and simple. Clearly, this is how he wants to be seen, and AI is a part of the effort to normalize this kind of image.
The second image — of an immigrant being detained by a US officer — is even more disgusting. Portrayed in the “Studio Ghibli” art style, this piece is another attempt to downplay the actual horror of the image at hand. Packaging this moment in an approachable, animated format warps a heartrending act of brutality into something less serious or important. The image is extra revolting when one considers that its purpose is to mock its subject — to create some kind of twisted fun off of someone’s life being destroyed. The way AI was used to create this image is a perversion of the creative spirit, and, as Hayao Miyazaki would say, “an insult to life itself.” This image is more than just vile, it is sinister. It co-opts the art style of a famously anti-fascist director to advertise a dystopian system of deportations. And yet, there’s no way for Miyazaki to protest his art style being used in this manner. In fact, there’s been little legal action against AI at all.
The fact of the matter is that AI is not regulated nearly enough in the United States. There is no comprehensive federal law which limits AI — and as such its use is dependent upon a variety of state and local laws. President Trump has even repealed the few federal attempts to place limits on AI, including an Executive Order from former President Biden that sought to label AI content to better combat the spread of misinformation. Allowing AI companies to run amok creates countless problems ranging from transparency to civil rights issues, and yet the President has no interest in holding them back. While regulation may slow innovation in this field, it is caution that should be our priority, not the unchecked growth of untrustworthy generative AI models. The proliferation of AI tools only serves the President’s aims, as they allow him to indiscriminately slash the federal workforce and surveil the American citizenry.
So, what is to be done? With the AI industry seeking to put artists out of work and profiting massively from the surveillance of Americans, is there any way to fight back? The short answer is yes. Firstly, the best way to get a message through to a corporation is to hit them where it hurts: in the profits. Boycott the AI of any company that contributes to this fascist project — wherever that AI may be found. It is our responsibility not to aid these companies in any way, and if that means boycotting something else that uses that AI, so be it. If companies see that they are losing money, there is a chance that they will rethink their relationship with Donald Trump.
Boycotting isn’t a silver bullet, by any means. It seems unlikely that Trump will pass any legislation curbing the influence of AI, but once his political power wanes, we must not forget to address this issue. AI has the potential to do great good, but it is currently being used to spread disinformation, to propagandize a fascist President, and to spy on the people the government is supposed to protect. AI needs strict guidelines for what it can and can’t be used for, and any content created by it needs to be marked as such. Time is of the essence, and we must act before our lives are controlled by something out of Orwell.
It may already be too late.
Featured Image Source: @WhiteHouse on X
Comments are closed.