Microsoft Released Bing Chatbot Despite OpenAI’s Warning about Its Unpreparedness

According to a report by The Wall Street Journal, the partnership between Microsoft and OpenAI has been strained due to tension and confusion. This has resulted in Microsoft’s internal AI team grappling with budget cuts and limited access to OpenAI technology. This confusion arose during the controversial roll-out of AI-powered Bing search, which was found to be susceptible to prompt injection attacks. OpenAI had warned Microsoft that integrating its technology with Bing would require more training. This was particularly regarding Bing’s chatbot, Sydney, which OpenAI feared would produce inaccurate or unhinged responses. Despite the warning, Microsoft launched the Bing chatbot, which attracted some criticism. Microsoft CEO, Satya Nadella, claimed that the initial hiccups were all part of the plan to train Sydney to respond to real-world prompts. However, unnamed sources reported that some Microsoft executives had concerns about the launch’s timing. They noted that OpenAI had started testing ChatGPT – which eventually became a runaway success – while Microsoft was still working on integrating OpenAPI tech into Bing. The release of Bing did not come close to the success of ChatGPT, which has acquired 200 million monthly users, and Bing has only 100 million daily active users. Bing and ChatGPT’s chatbots were later criticised for making up stuff, lying, and other issues. OpenAI is still looking for ways to limit chatbots’ hallucinations. While Bing responded with fewer unhinged outbursts during long conversations, it continues to get things wrong.