ChatGPT creator OpenAI is testing how to get broad feedback on decisions that affect its AI, its president Greg Brockman said Monday.
At the AI Forward event in San Francisco, hosted by Goldman Sachs Group Inc and SV Angel, Brockman discussed the broad outlines of how the maker of the wildly popular chatbot is working to regulate AI globally.
One of the announcements he previewed is reminiscent of Wikipedia’s model, where he said people with different views come together and agree on encyclopedia entries.
“We don’t just sit in Silicon Valley and think we can write these rules for everybody,” he said of AI policy. “We are starting to think about democratic decision-making.”
Another idea discussed by Brockman, which OpenAI elaborated on in a blog post on Monday, is that governments around the world should coordinate efforts to ensure the safe development of artificial intelligence.
Since its launch on November 30, ChatGPT’s generative AI technology, capable of spinning insanely authoritative prose from text prompts, has captivated audiences and made the program the fastest-growing app of all time. AI has also raised concerns about its ability to create deep fake images and other misinformation.
In evaluating the development of artificial intelligence, Brockman looked at Wikipedia and elsewhere. He and OpenAI said a body like the International Atomic Energy Agency (IAEA) could set limits on deployment, enforce compliance with safety standards and monitor computing power usage.
Another proposal was a global agreement to limit the annual growth of border AI capabilities, or a joint global project in which major governments could participate.
OpenAI CEO Sam Altman last week proposed various ideas to U.S. lawmakers to put AI safeguards in place, including requiring licenses to develop the most advanced AI models and establishing an associated regulatory system. He will visit European decision-makers this week.
Read all the Latest Tech News here.