Sen. Bernie Sanders presses AI assistant Claude on pervasive data collection, backs moratorium on new AI data centers

Sen. Bernie Sanders asked the AI assistant Claude how much information AI systems collect and what would surprise Americans about the practice, saying he has been examining AI’s effects on jobs, children’s mental health and the political process.

Claude responded: "Companies are collecting data from everywhere. Your browsing history, your location, what you buy, what you search for, even how long you pause on a web page." Claude said companies combine thousands of data points into detailed profiles and use those profiles to determine "what ads you see, what prices you're shown, even what information gets prioritized in your social media feed," and called the activity "invisible and largely unregulated."

Sanders pressed on motive, asking why so much information is collected. Claude replied bluntly: "Money, Senator," saying companies use AI to turn data into valuable insights for targeted advertising and differential pricing. Claude added that data brokers buy and sell information "about millions of Americans without their knowledge," and that attention and behavior have become commodities.

Turning to politics, Sanders asked how profiling affects elections and public discourse. Claude warned that AI-enabled microtargeting can "identify voters based on incredibly specific vulnerabilities" and enable campaigns to show "completely different narratives to different groups," fragmenting shared information and undermining democratic processes. Claude also warned that if "bad actors, foreign governments for instance, get access to these profiles," they could use them to "sew division and manipulate elections."

Sanders asked whether the public can trust AI companies to protect privacy when those companies use personal information to train models. Claude described an inherent conflict of interest: companies that profit from personal data have limited incentive to limit its use, and "there's almost no accountability" without legal safeguards. Claude recommended "transparency requirements, consent rules, penalties for misuse" and rights for people to access and delete their information.

Sanders asked whether a moratorium on the development of new AI data centers would be appropriate to slow expansion while safeguards are put in place. Claude said that while a moratorium "would buy us time to put guard rails in place," it also carries tradeoffs because it would pause innovation; Claude advocated a more targeted approach—explicit consent, limits on training data, and transparency—while emphasizing speed: "What matters most is that we act fast because everyday companies are collecting more data and building more capabilities."

Sen. Sanders responded that AI companies are "pouring hundreds of millions of dollars into the political process" to block safeguards, making the targeted-approach unlikely to succeed. In light of that political reality, Claude acknowledged the argument: "A moratorium on new data centers is actually a pragmatic response to that problem. It forces a pause that gives lawmakers like yourself actual leverage to demand real protections before companies can keep expanding."

Claude closed by framing privacy as a democracy issue: when organizations hold detailed profiles of millions of people they can "manipulate your choices, predict your behavior, and influence your thinking." The exchange ended with mutual thanks.

The exchange recorded no formal proposals, votes or legal action; it presented policy options — stricter privacy rules, transparency and consent requirements, and a possible moratorium on new AI data centers — for lawmakers to consider.