AI Business虎嗅

Sam Altman Interview: I Want the US to Lead, But I Oppose Only Americans Benefiting

虎嗅Oct 14, 20261 min read

AI Summary

In a deep interview with POLITICO's Decoded column, Sam Altman identified two ways AI could go wrong: loss of control, and excessive power concentration in the name of safety. He acknowledged major disagreements with Anthropic, arguing society should accept limited, understandable harms but never catastrophic risk, and opposed licensing for models one or two generations behind the frontier. He also addressed the White House's superintelligence executive order and OpenAI's stance on the recent safety commitments.

AI Insights

Altman's framing positions OpenAI as a pragmatic middle path between uncontrolled AI risk and dangerous power concentration, a stance that could shape how regulators distinguish frontier-model oversight from broad licensing. His public split with Anthropic over accepting limited harms in exchange for wide access signals that the industry's governance debate is now a competitive as well as a safety question. Expect this to influence the design of voluntary commitments and any future licensing thresholds tied to model capability.

Analysis written by AI from the original article. Check it against the source.

Translated text

Sam Altman said AI could go wrong in two ways: loss of control, and excessive concentration of power justified by safety. OpenAI aims to be a pragmatic middle ground between these extremes. He acknowledged a major gap with Anthropic. He said that a scenario where a single lab in San Francisco controls the technology and distributes benefits from the top down is an unacceptable trade-off for human freedom and autonomy. He argued that society should accept limited, understandable harms in exchange for the benefits, but catastrophic risks must never occur. On regulation, he supports testing requirements only for frontier models and opposes licensing for models one or two generations behind. He also said OpenAI is not a 'doomer' lab and has been the most determined to release the technology to the public. On the White House's plan to rename 'artificial intelligence' as 'superintelligence,' he said the term is more accurate for the coming changes, but people will likely keep calling it AI.

Related News

AI BusinessIT之家

Refusing the Ruling: Anthropic Pushes Back Against US Government's AI 'Supply Chain Risk' Designation

Anthropic has asked a US federal appeals court to reconsider a ruling that upheld the Trump administration's designation of the AI company as a supply chain risk to national security. The company argues the panel misinterpreted the law and that the designation is retaliation for its public statements on AI safety. The case will help determine whether the federal government or private companies control safety safeguards for AI technology.

AI BusinessIT之家

OpenAI Saudi-Born Engineer Arrested by FBI for Secretly Installing Recording Device in Executive Conference Room

According to the U.S. Department of Justice, the FBI arrested Abdulrahman Alfozan, a Saudi-born engineer at OpenAI, on charges of secretly installing a USB-sized recording device in an executive conference room at the company's San Francisco office. OpenAI fired the employee and referred the case to federal investigators.

AI Business虎嗅

AI Personal Assistants, AI Phones and Embodied Intelligence Have Yet to Find Real Demand

A Chinese investment columnist outlines Zeng Ming's three-stage model of AI: infrastructure, mass application, and AI-native applications. He argues that only AI Q&A and AI programming have truly monetized so far, while AI personal assistants, AI phones and embodied intelligence remain driven by corporate strategic ambitions rather than genuine consumer demand.