May 16, 2023

Altman: AI can go quite wrong

OpenAIAn AI research company — created GPT, ChatGPT, DALL-E, and the o-series reasoning models. CEO Sam Altman tells the Senate Judiciary Committee that frontier AI needs government oversight — while arguing for licensing and safety testing.

Quote

If this technology goes wrong, it can go quite wrong, and we want to be vocal about that.

What it was for

Sam AltmanAltman walked a tightrope: urging regulationGovernment rules governing technology — privacy, competition, safety, and export controls. that could entrench incumbents while presenting OpenAIAn AI research company — created GPT, ChatGPT, DALL-E, and the o-series reasoning models. as the responsible actor. The line became the sound bite from the first major congressional hearing on generative AIModels that create new content — text, images, audio, or code — from learned patterns..

People

  • Sam Altmanceo

Companies

  • OpenAI

Why it's here

The defining quote from Congress's first ChatGPTOpenAI's conversational chat product — brought large language models to hundreds of millions of users.-era AI hearing.

Why it mattered

It mainstreamed AI risk talk in Washington months after ChatGPTOpenAI's conversational chat product — brought large language models to hundreds of millions of users.'s launch.

What it solved

Lawmakers needed a vocabulary for catastrophic AI failure from industry itself.

Media

  • YouTubeCNBC — OpenAI CEO Sam Altman Senate AI oversight testimony (May 16, 2023)
  • Sam Altman
    ImageSam Altman

    首相官邸ホームページ, CC BY 4.0, via Wikimedia Commons

Related