OpenAI, the creator of the wildly popular ChatGPT chatbot, is exploring ways to gather broad input on decisions impacting its artificial intelligence. At AI Forward, an event hosted by Goldman Sachs and SV Angel, OpenAI President Greg Brockman discussed the broad contours of how the company is seeking regulation of AI globally.
Brockman proposed a model similar to Wikipedia, which requires people with diverse views to coalesce and agree on entries. He also suggested that governments around the world should coordinate to ensure AI is developed safely, and proposed the idea of a global agreement to limit the annual growth of frontier AI capabilities.
OpenAI CEO Sam Altman has also proposed various ideas to U.S. lawmakers last week for setting guardrails for artificial intelligence, such as requiring licenses to develop the most sophisticated AI models and establishing a related governance regime. He is visiting European policymakers this week to discuss these ideas.
The rapid growth of AI has raised concerns over its ability to create deepfake pictures and other misinformation. OpenAI’s efforts to seek broad input on decisions impacting its AI is a step in the right direction to ensure the technology is developed safely and responsibly.