OpenAI CEO Sam Altman appeared before the United States Senate Committee on the Judiciary and answered questions from lawmakers about the potential unintended consequences of generative artificial intelligence, Gizmodo reports.
He urged lawmakers to pass new AI laws and regulations as soon as possible to potentially set standards for technology companies to train and release new AI systems. According to Altman, regulatory intervention by governments will be critical to reducing the risks associated with increasingly powerful AI models.
Along with other experts, he also noted that ChatGPT-style coordinated models, if left unregulated, could increase the amount of misinformation on the Internet, facilitate cybercriminals, and even threaten the credibility of election systems.
With that in mind, Altham and lawmakers explored numerous potential solutions and considered creating a new regulatory body of AI experts to oversee the development and implementation of the technology.
Altman also acknowledged another concern during the hearing — the potential for AI to trick users with sophisticated chatbots is growing rapidly. At the same time, he compared it to the situation that society faced when Adobe’s Photoshop appeared.
“I think people are able to adapt quickly,” Altman said. “When Photoshop came onto the scene a long time ago, for a while people were really quite fooled by photoshopped images but then pretty quickly developed an understanding that images might be photoshopped. This will be like that but on steroids.”
Connecticut senator Richard Blumenthal also warned about the dangers of unregulated AI during the hearings. To do this, he used a recording of a voice cloning program reading a prompt generated by ChatGPT.
“This is not the future we want,” the AI-generated Blumenthal impersonator said.
Many lawmakers, including Sen. Blumenthal and Sen. Cory Booker of New Jersey, criticized their colleagues for losing the initiative when it came to preventing social media from becoming a vehicle for spreading misinformation and harmful content. Artificial intelligence can potentially be an opportunity to remedy the situation.
“Congress failed to meet the moment on social media now we have the obligation to do it on AI before the threats and risks become real,” Blumenthal said.
But Missouri Senator Josh Hawley believes that large language models like GPT4 could go in one of two very different directions.
“We could be looking at one of the most significant technological innovations in human history,” Hawley said. “My question is what kind of innovation is it going to be? Is it going to be like the printing press that diffused knowledge and power and learning…or is it going to be more like the atom bomb.”
At the same time, Tennessee Senator Marsha Blackburn sparked a debate about AI’s use of copyrighted material. She shared how an AI trained on country singer Garth Brooks was able to create a convincing version of his song. The senator used this example to demand compensation from Altman’s company if the artists’ works help the AI model in training.
It was previously reported that artificial intelligence is capable of changing users’ beliefs without their knowledge. This is evidenced by the growing number of studies in this direction.