“We are the AI.” That Borg-like greeting could be coming soon to the internet in the form of new AI overlords. In a recent chilling interview, Microsoft founder and billionaire Bill Gates called for the use of artificial intelligence to combat not just “digital misinformation” but “political polarization.”
He is only the latest to call for the use of either AI or algorithms to shape what people say or read on the internet. The danger of such a system is evident where free speech, like resistance, could become futile.
In an interview on a German program, “Handelsblatt Disrupt,” Gates calls for unleashing AI to stop certain views from being “magnified by digital channels.” The problem is that we allow “various conspiracy theories like QAnon or whatever to be blasted out by people who wanted to believe those things.”
Gates added that AI can combat “political polarization” by checking “confirmation bias.”
Confirmation bias is a term long used to describe the tendency of people to search for or interpret information in a way that confirms their own beliefs. It is now being used to dismiss those with opposing views as ignorant slobs dragging their knuckles across the internet — people endangering us all by failing to accept the logic behind policies on COVID, climate change or a host of other political issues.
This is not the first call for AI overlords to protect us from ourselves. Last September, Gates gave the keynote address at the Forbes 400 Summit on Philanthropy. He told his fellow billionaires that “polarization and lack of trust is a problem.”
The problem is again … well … people: “People seek simple solutions [and] the truth is kind of boring sometimes.”
Not AI, of course. That would supply the solutions. Otherwise, Gates suggested, we could all die: “Political polarization may bring it all to an end, we’re going to have a hung election and a civil war.”
Others have suggested a Brave New World where citizens will be carefully guided in what they read and see. Democratic leaders have called for a type of “enlightened algorithm” to frame what citizens access on the internet. In 2021, Sen. Elizabeth Warren (D-Mass.) objected that people were not listening to the informed views of herself and leading experts. Instead, they were reading views of skeptics by searching Amazon and finding books by “prominent spreaders of misinformation.”
Warren blamed Amazon for failing to limit searches or choices: “This pattern and practice of misbehavior suggests that Amazon is either unwilling or unable to modify its business practices to prevent the spread of falsehoods or the sale of inappropriate products.” In her letter, Warren gave the company 14 days to change its algorithms to throttle and obstruct efforts to read opposing views.
Social media responded to such calls and engaged in widespread censorship of those who held opposing views of mask mandates, vaccine safety, school mandates, and the origin of COVID-19. Many of those criticisms and views are now acknowledged as plausible and legitimate, but scientists were banned and censored. There was no “polarization” allowed. The public never was allowed to have that full debate on social media because such views were declared disinformation.
President Biden joined in these calls for censorship, often sounding like a censor-in-chief, denouncing social media companies for “killing people” by not blocking enough. Recently, he expressed doubt that the public can “know the truth” without such censorship by “editors” in Big Tech.
They found an eager body of censors at companies like Twitter. After taking over as CEO, Parag Agrawal pledged to regulate content as “reflective of things that we believe lead to a healthier public conversation.” Agrawal said the company would “focus less on thinking about free speech” because “speech is easy on the internet. Most people can speak. Where our role is particularly emphasized is who can be heard.”
That view was echoed last week in the first hearing on Twitter’s censorship program. Former Twitter executive Anika Collier Navaroli testified on what she repeatedly called the “nuanced” standard used by her and her staff on censorship. She explained that they did not just balance free speech against public safety in deciding whether to allow someone to speak. Rather censorship depended on the persons involved: “Whose free expression are we protecting at the expense of whose safety and whose safety are we willing to allow to go the winds so that people can speak freely?”
All of that could be much easier with an AI Overlord that can protect us against our own doubts and divisions. Currently, Microsoft, the company Gates founded, uses NewsGuard, a self-described arbiter of misinformation, which rates sites and has been widely criticized for targeting conservative media.
Now, this work could be turned over to an AI Overlord. Of course, the intelligence remains artificial. A human has to program what is truth and what is intolerable “polarization.” It would be a ramped-up version of ChatGPT, the popular AI service that Microsoft just incorporated into its Bing search engine. It censors “offensive” content and bars certain viewpoints because it was told to do so.
AI enforces the collective truth that needs to be amplified for a greater good as determined by figures like Gates.
You would be forgiven if it all sounds chillingly similar to “Lower your shields and surrender … Resistance is futile.”
Jonathan Turley is an attorney and a professor at George Washington University Law School.