• Source:JND
Add as a preferred

Generative AI, which has been in the spotlight since OpenAI released its capable chatbot ChatGPT to the public last year, is back in the spotlight as the ChatGPT creator has cautioned Microsoft to slow down in integrating its GPT-4 model into its services because of its ambiguous and imprecise responses.

While Microsoft was preparing to create an AI-powered Bing using the GPT 4 model, ChatGPT developer Sam Altman cautioned Satya Nadella's tech powerhouse to proceed slowly with the GPT-4 integration. According to the Wall Street Journal, Microsoft continued with the integration of GPT-4 technology into Bing "despite cautions that it might take time to minimise the inaccurate and peculiar responses."

Microsoft and OpenAI allegedly disagreed on generative AI as the report states that OpenAI cautioned about the dangers of prematurely releasing a chatbot based on an unreleased version of GPT-4.

"The rollouts of ChatGPT last fall and Microsoft's AI-infused Bing months later also created tension," according to a report. Microsoft staff were also reportedly concerned that "ChatGPT would steal the thunder from the new Bing." While some countered that Bing AI may benefit from the knowledge gained from the widespread use of ChatGPT.

Users noticed inaccurate replies and worries regarding tool interactions after the Bing Chat was introduced at the beginning of this year. Microsoft immediately restricted Bing Chat responses to prevent the AI from acting strangely, and they continued to reduce the number of replies and session length in accordance with the most recent advancements.

Meanwhile, the business has introduced GPT 3.5 turbo and GPT 4, two new language model iterations. The platform can now allow call functions, and the company has further revealed that it would let programmers specify programming functions and have the models generate the necessary code to carry them out.