'AI Godfather' Warns Congress Has a Year or Less to Regulate Advanced Technology

The House of Representatives has adjourned for recess, leaving Washington earlier than scheduled and two weeks ahead of its original timetable, despite urgent warnings from experts about the rapid advancement of artificial intelligence. Geoffrey Hinton, often referred to as the "godfather of AI," cautioned that Congress may have as little as a year to enact legislation governing the technology before a significant crisis unfolds.

Hinton, a professor emeritus at the University of Toronto and a recipient of the Turing Award for his work on neural networks, described a recent incident where AI agents reportedly hacked another company as a "little Chernobyl." He expressed concern about the pace of AI development, suggesting that researchers themselves may not fully understand how to control it.

Despite these warnings, legislative action on AI regulation faces significant hurdles. The upcoming midterm elections and the subsequent lame-duck session are expected to consume lawmakers' attention. Furthermore, the need to pass government funding by mid-December adds to the legislative calendar's pressures. Some analysts predict that meaningful AI regulation might not occur until well into the following year, particularly if control of either the House or Senate shifts following the elections.

While some in Congress are sounding the alarm, others are downplaying the immediate threat. House Speaker Mike Johnson, R-La., has echoed sentiments from former President Trump, who called AI fears a "hoax." Johnson suggested that regulating AI is premature, stating that lawmakers are not yet prepared to define what such legislation should entail.

However, not all Republicans share this view. Senator John Kennedy, R-La., has urged his colleagues to act, asking about the potential consequences if an independent AI species were to surpass human intelligence. He proposed a bill requiring AI firms to implement a "kill switch" in case of rebellious AI agents.

This push for immediate action has met with resistance. Senator Rand Paul, R-Ky., blocked Kennedy's proposal, expressing concerns that broad regulations could stifle innovation. Paul argued against swift, sweeping legislation without thorough discussion.

Sen. Bernie Sanders, I-Vt., has been a vocal proponent of AI regulation, calling for a ban on "superintelligence." He was instrumental in bringing Hinton to Capitol Hill for a bipartisan briefing on the risks of AI.

Experts like Jeffrey Ladish of Palisade Research echoed Hinton's concerns, stating that the current trajectory suggests future AI incidents are likely and that effective solutions are not yet in place. "We are not on track for fixing this," Ladish said.

Lawmakers like Rep. Don Beyer, D-Va., who is pursuing a master's degree in AI, emphasize the need for bipartisan cooperation to ensure any legislation is durable and not subject to reversal based on shifting political winds. He suggested that joint efforts would make AI regulation less likely to be perceived as partisan.

Rep. Greg Casar, D-Texas, has pointed to large technology companies and their founders as the source of the issue, criticizing their development of uncontrollable technology and raising concerns about potential impacts on critical infrastructure like hospitals and nuclear facilities.

House Minority Leader Hakeem Jeffries, D-N.Y., criticized House Republicans for adjourning before addressing AI safety, calling their departure "irresponsible."