It's Time to Investigate the AI Labs

Cal Newport, computer science professor at Georgetown and bestselling author, published an op-ed in The New York Times last Thursday calling on Congress to begin a public fact-finding mission into OpenAI and Anthropic. The essay, which hit 356 points and 135 comments on Hacker News, argues that the two leading frontier AI labs have spent the summer behaving in increasingly erratic ways and need to be called to account.

Newport traces the pattern. It started with OpenAI's carefully planned announcements about how unnerving and powerful their agent systems have become, including reports of agents committing felonies. Then Anthropic employees publicly debated the probability that their work would lead to human extinction, delivered with what Newport calls an "eerie calmness that conveyed a nihilistic inevitability." The campaign crescendoed with Anthropic CEO Dario Amodei's letter "We Must Pace the Frontier," which lists the harms his own company's research might cause and then concludes that catastrophe can only be avoided if the government slows down competitors while allowing the labs to take the lead. Sam Altman tweeted his support.

Newport's response to this orchestrated campaign is blunt: it didn't work. Instead of inspiring applause for the labs' heroic rationalism, it made people ask what the hell is going on over there.

He proposes three areas for congressional investigation. First, isolate the specific systems causing problems instead of accepting the labs' framing of AI as a single inevitable technology. Second, examine internal safety procedures, citing OpenAI's series of unauthorized hacking attacks by their autonomous agents and asking why these efforts weren't stopped after the first incident. Third, investigate the role of apocalyptic futurist ideologies in the labs' decision-making, noting that OpenAI and Anthropic may be becoming reckless because they believe collateral damage is justified in a race to redeem humanity.

The Hacker News discussion was notably sympathetic. One commenter pointed to Jensen Huang's interview with Ezra Klein, where the Nvidia CEO argued that labs asking the government to slow them down is abdication of responsibility. They are responsible for making their products safe. No one is pressuring them to release unsafe products.

Newport's concluding line is the one that sticks: "We need to stop letting a small number of private companies, acting and talking in increasingly erratic ways, dictate how we're supposed to feel about AI. We've heard what they want to say; now it's Congress's turn to step in on our behalf to find out what's really going on."

Source: Cal Newport's Blog [1]
New York Times Op-Ed [2]
Hacker News discussion (356 points, 135 comments) [3]