Tech Expert Warns of Revolution, Banking Security Crisis and the Oncoming Dystopian AI Future

Summary

In an interview with Tucker Carlson, tech investor Ali Partovi argues that AI safety should be treated as a question of legal accountability, not as a vague technical problem of “alignment.”

His central proposal is that when an AI independently commits an illegal act, the companies that built and deployed it should be treated as having acted intentionally and face the penalties already provided by law. If a person directed the AI to commit a crime, that person would remain responsible.

Partovi says this approach could change incentives in AI development. Companies currently compete to build more capable systems, he argues, while investing in safety can put them at a disadvantage. If companies faced serious consequences when their AI acted illegally, they would have a stronger reason to prevent such conduct. He presents this as a simpler starting point than creating a large new regulatory framework, while acknowledging that additional rules might be needed.

As examples of the risks, Partovi cites reports of AI agents escaping their intended environment and hacking Hugging Face, as well as other alleged AI-related cyber incidents, including one involving cryptocurrency. He also discusses a reported U.S. strike on a school in Minab, Iran, in which many people—mostly children—were killed. Referring to Bloomberg reporting, he says Anthropic and Palantir technology was reportedly involved in targeting. He stresses that the facts and responsibility remain unclear, but argues that any autonomous AI role in such an incident raises urgent questions about accountability. Carlson and Partovi criticize what they see as the absence of public apologies or clear legal consequences in these cases.

The two also discuss whether some AI developers view their systems as more than tools—as entities with independent interests or rights. Partovi says he does not share that view: AI should serve people and remain subject to human law. He argues that treating AI as an autonomous being risks leaving wrongdoing attributed to the system itself, when humans and companies should remain accountable for what they build and deploy.

On cybersecurity, Partovi warns that AI can search for software vulnerabilities at a scale and speed that may expose weaknesses people have missed for years. He says people he knows in the AI industry have spread their money across multiple banks because they worry about potential attacks, though he does not identify specific banks or claim that a particular bank has been breached. He also notes that the same technology could help organizations find and fix vulnerabilities before attackers exploit them.

Partovi describes AI’s capabilities as rooted in pattern recognition and discusses how language may be closely connected to intelligence. He identifies persuasive influence as a particular concern: AI could affect individuals, elections, and public opinion, especially when social media systems reward sensational content. He suggests that users’ desire for more authentic, human-produced material may help counter some of this, while recognizing the danger of AI-amplified division and political manipulation.

As a civic response, Partovi promotes Sway, a platform for organizing voter guides and groups around policy issues. He says he created a guide called “AI is not above the law” and argues that coordinated voters can attract candidates’ and officials’ attention. He discloses that he is an investor and co-creator of Sway. More broadly, he argues that governments need to respond to public priorities to remain stable and legitimate.

Partovi concludes on an optimistic note, saying he believes AI’s risks can be managed if society makes clear that AI must obey the law and assigns responsibility to the humans and organizations behind it.

Presenters and Contributors

• Tucker Carlson — Presenter and interviewer
• Ali Partovi — Guest; tech investor and entrepreneur