- OpenAI CEO Sam Altman testified before Congress, advocating for government regulation of AI systems, emphasizing the need for collaboration between the government and tech companies. He noted the potential harm of AI misuse and the importance of the technology being transparent to users.
- Altman, creator of AI text-prompt tool, ChatGPT, reassured lawmakers that OpenAI had conducted thorough evaluations to prevent misuse of their technology. He proposed a regulation approach that includes licensing and testing requirements for AI models exceeding certain capabilities.
- The hearing occurred amid increased public attention towards AI, heightened by the recent launch of ChatGPT 4 and Google’s competitor, Bard. The testimony underscored the necessity of lawmakers' understanding of AI complexities for effective regulation, potentially a significant challenge given previous difficulties with technology legislation.
Artificial intelligence technology has captured the attention of policymakers, prompting discussions about regulation.
Sam Altman, the CEO of OpenAI, testified before Congress, urging the government to take action in regulating AI systems. Altman's testimony aligns with the growing concerns surrounding AI's potential harms and the need to prevent its misuse. In his address, Altman emphasized the importance of collaboration between the government and tech companies in drafting this type of legislation.
Why was he asked to testify? He’s the creator of ChatGPT, the game-changing AI text-prompt tool.
An appearance before Congress marked a significant milestone for the Stanford dropout turned tech entrepreneur. Before OpenAI, Altman was president of Y-Combinator, a startup accelerator that has helped launch companies like Airbnb, Coinbase, Reddit, and Twitch.
During his testimony, he underscored the potential consequences if AI technology goes awry.
“I think if this technology goes wrong, it can go quite wrong,” he said.
Altman's plea for regulation resonated with lawmakers who demonstrated a limited understanding of AI's capabilities and risks during the hearing.
The hearing came just weeks after a meeting at the White House involving prominent tech CEOs, including Altman. With the release of ChatGPT 4 and Google’s competitor, Bard, AI has been at the forefront of the public’s mind.
The Testimony
His opening statement:
Sam Altman's opening statement
AI has the potential to improve nearly every aspect of our lives, but also that it creates serious we have to work together to manage. We think it can be a printing press moment. We have to work together to make it so. OpenAI is an unusual company and we set it up in that way because AI is an unusual technology. We are governed by a non-profit and our activities are driven by our mission and our charter, which commits us to ensure the broad distribution of the benefits of AI and to maximizing the safety of AI systems. We are working to build tools that can one day help us make discoveries to address some of the world’s biggest challenges like climate change and curing cancer. Our current systems aren’t yet capable of doing these things but it has been immensely gratifying to watch many people around the world get so much value from what these systems can already do today.
Senator Richard Blumenthal challenged Altman on some of the negative consequences of AI, including deepfakes, voice impersonations, and potential job losses.
“Congress has a choice now. We had the same choice when we faced social media we failed to seize that moment. The result is predators on the internet, toxic content, exploiting children, creating dangers for them.”
Senator Josh Hawley tried to draw a comparison to other technologies.
“Is it going to be like the printing press, that diffused knowledge and power and learning widely across the landscape,” he asked, “or is it going to be more like the atom bomb, a huge technological breakthrough but the consequences severe, terrible, continue to haunt us to this day.”
Altman responded with reassurance that OpenAI had conducted extensive evaluations, external red-teaming, and dangerous capability testing.
He also encouraged government regulation.
“We think that regulatory intervention by governments will be critical to mitigate the risks of increasingly powerful models.”
He offered a few approaches the Senate might consider, like a combination of licensing and testing requirements for the development, and release of AI models above a threshold of capabilities.
Christina Montgomery, Vice President and Chief Privacy & Trust Officer for IBM, was also testifying at the hearing. She encouraged Congress to adopt a precision regulation approach, establishing rules based on specific use cases, not regulating the technology itself.
She outlined four guidelines for this approach:
- Different rules for different risks. The strongest regulation applied to use cases with the greatest risks.
- Clearly defining risks.
- AI shouldn't be hidden. Consumers should know when they're interacting with an AI system.
- For higher-risk use cases, companies should be required to conduct impact assessments that show how their systems perform against tests for bias.
Altman echoed point number three, saying that it should be obvious to users when they are interacting with AI.
When pressed again about jobs disappearing, Altman offered an optimistic glimpse into the future, saying that he believes there will be far better jobs on the other side of AI development, and that jobs currently in existence will get better.
He also emphasized that GPT4 cannot take over a job at the moment, but can help people do their job.
“GPT4 is good at doing tasks, not jobs,” he said.
Senator Hawley followed up, asking Altman about the potential for one of these models to be used in a disinformation campaign to affect an election. He responded in the affirmative, saying one of his biggest concerns is the future ability of AI to “manipulate and persuade.”
“Given that we're going to face an election next year and these models are getting better, I think this is a significant area of concern... I do think some regulation would be quite wise on this topic.”
Altman took an apparent dig at Google, and other ad-based revenue companies:
“We don't have an ad-based business model so we're not trying to build up these profiles of our users. We're not trying to get them to use it more, actually, we'd love it if they use it less because we don't have enough GPUs. But I think other companies are already, and certainly will in the future, use AI models to create very good ad predictions of what a user will like.”
There was widespread agreement that there needed to be a separate agency to regulate AI, even from conservatives. Senator Lindsey Graham asked the panel:
“Do you agree with me that the simplest way and the most effective way is to have an agency that is more Nimble and smarter than Congress?”
Christina Montgomery, IBM’s representative, believed that AI should be regulated using the existing systems in place already, not with an entirely new agency.
Gary Marcus, a professor and cognitive scientist also testifying, agreed that there should be a separate agency and even global AI regulation laws.
In a surprising twist, Altman confirmed he has no equity in OpenAI, which owes nearly the entirety of its $29 billion net worth to ChatGPT.
Senator John Neeley Kennedy bantered with him on his financial situation. “You make a lot of money, do you?”
“Enough for health insurance. I own no equity in OpenAI,” he responded.
“Really?” Kennedy asked. “That’s interesting. You need a lawyer”.
“I’m doing this because I love it,” he responded.
This is INSANE.
— Sam Parr (@thesamparr) May 16, 2023
Sam Altman saying he owns 0% equity in OpenAI.
Surely there's something more to this story.
Otherwise, @sama is just a very, VERY unique person. https://t.co/sOi6JMpSe2
Altman’s net worth is estimated to be $500 million, made from investments during his time at Y-Combinator.
The Challenges of AI Legislation on Capitol Hill
The path to AI regulation is not without challenges. There is significant disagreement in the tech industry for how AI should be regulated, if at all— and the concerns voiced during Altman's testimony mirror the previous Congressional battles over social media policies. Republicans express worries about the potential censorship of conservatives, while Democrats are concerned about hate speech and disinformation. These differing viewpoints pose a challenge to finding common ground and formulating bipartisan AI policies.
Like with social media, there is a knowledge gap among lawmakers regarding the complexities of AI technology. This was highlighted by their reference to Section 230, a provision that protects online platforms from legal liability for user-generated content. Altman tried to emphasize the distinction between social media and the subject of the hearing.
“I think it's tempting to use the frame of social media. But this is not social media. This is different and so the response that we need is different.”
Section 230 is 27 years old, which is ancient for internet technology. Critics argue that this broad immunity has allowed platforms to evade accountability for harmful content, while supporters maintain that it promotes free expression and innovation. Congress has been unable to agree on an effective update to the provision, suggesting AI regulation could be years away.
ChatGPT is only a specific implementation of AI technology. To effectively regulate AI, policymakers must acquire a deeper understanding of its intricacies and potential applications, beyond the scope of text-prompt tools like ChatGPT.
Is it Alive?
Altman explained his conception of ChatGPT.
“I think it's important to understand and think about GPT4 as a tool, not a creature, which is easy to get confused. And it's a tool that people have a great deal of control over.”