Greg Brockman, president, and cofounder of OpenAI admitted the company “made a mistake” in an interview with The Information in response to Elon Musk’s criticism of ChatGPT for being too woke.
Musk, a former cofounder of OpenAI who has since cut connections with the firm, frequently chastises OpenAI for putting in place measures that stop the chatbot from delivering potentially harmful responses.
Everything you need to know about ChatGPT’s criticism!
The chatbot declined to produce a favourable poem regarding Donald Trump last month, claiming it wasn’t trained to produce “partisan, biassed, or political” content. Screenshots of the interaction were shared on Twitter. Nevertheless, when given the identical assignment but with Joseph Biden instead of Trump, the chatbot produced a beautiful poem. The chatbot’s refusal to produce a poem about Trump was deemed “a big worry” by Musk.
The technology had previously come under fire from Elon Musk, who claimed that “the danger of educating AI to be woke that is, to lie is deadly.” As more users swarm to OpenAI-powered AI chatbots such as ChatGPT and Bing’s freshly launched chatbot, their limitations and shortcomings have come to light. In response, businesses have enhanced technology with guardrails.
Microsoft placed conversational restrictions on its AI-powered Bing chatbot in the month after it launched it, limiting users to a daily question maximum of 50 and a session question limit of five. Later, it loosened those restrictions. However, ChatGPT is a work in progress, and based on Brockman’s remarks, it appears that the platform will keep changing.
Also a work-in-progress, Brockman’s remarks imply that ChatGPT will continue to develop. The Information quoted Brockman as saying, “Our goal is not to develop an AI that is prejudiced in any specific direction. For the sake of all users, we want OpenAI’s default personality to be fair. It’s difficult to operationalize exactly what that means, so I don’t think we’ve completely arrived.”