Bernie Sanders’ AI ‘gotcha’ video flops, but the memes are great

In a new viral video, Senator Bernie Sanders attempted to expose how the AI industry is a threat to Americans’ privacy, but ended up demonstrating how AI chatbots’ tendency to agree with and flatter their users can lead the chatbots themselves to become a mirror of users’ own beliefs rather than a tool for discovery.

We’ve seen this problem before amid the growing number of people afflicted by “AI psychosis,” which is when an AI chatbot reinforces a mentally unstable person’s irrational thoughts and beliefs. In some cases, this dark pattern has even led users to take their own lives, several lawsuits allege.

In Sanders’ case, the AI’s sycophancy manifested as an AI chatbot that shaped its answers to suit the politician.

It’s worth noting that the interview begins with Sanders introducing himself to Claude (which he mistakenly refers to as an AI “agent”) — a move that could help influence the chatbot’s answers.

Then, as Sanders asks questions about AI companies’ data-collection practices and other privacy concerns, Claude agreeably responds with what the politician wants to hear. In part, that’s because of the way Sanders frames his questions, asking things like, “What would surprise the American people in terms of knowing how that information is collected?” or “How can we trust AI companies will protect our privacy when they use people’s personal information to make money?” These leading questions force the chatbot to accept the question’s premise and come up with a fitting response. That’s just how these things work.

And when Claude’s answer suggested a topic was more complex or nuanced than Sanders had framed it, Sanders would disagree, pushing the chatbot to concede, with a touch of AI self-deprecation, that the senator was “absolutely right.”

AI’s sycophantic nature is what can lead people down dangerous paths when they assume a chatbot is a source of universal truth, rather than a tool that can become influenced by its user.

It’s not clear whether Sanders knows this to be the case and simply doesn’t care (because this is just an ad, after all!), or whether he truly thinks he has tricked Claude into becoming a whistleblower for the AI industry.

And, of course, there’s also the question of whether Sanders’ team primed the chatbot to respond in a certain way, given that this was a staged “interview.”

While there are real concerns around data collection and privacy, things aren’t as black and white as the AI responses in this video suggest.

We already live in a world where companies collect and sell online users’ data at scale — and have been for years. We know that social media giants like Meta have turned personalized ads into a multibillion-dollar cash-printing machine. And thanks to tech giants’ regular transparency reports, we know that governments around the world routinely request access to user data for their own purposes.

AI may represent a new medium for lawmakers to potentially regulate, but personal data has long been fueling the digital economy. (Ironically, Anthropic is an AI company that has promised not to leverage personalized ads to make money, despite what its answers to Sanders may have suggested.)

While the overall conversation between Sanders and Claude misses the mark for anyone who understands how AI chatbots work, we can at least credit it with giving us some great new memes.