People with knowledge of its plans say that Microsoft Corp. is in talks to invest up to $10 billion in OpenAI, the company that made the artificial intelligence bot ChatGPT that went viral.
The software giant from Redmond, Washington, would put the money in over several years, according to the plan being considered. However, the final terms could change, said the people, who asked not to be named because they were talking about a private matter. They also said that the two companies had talked about the deal for months.
People familiar with the talks told Semafor that the possible investment would involve other venture capital firms and could put OpenAI’s value at about $29 billion. The documents sent to investors said that the deal would close by the end of 2022.
Representatives from both Microsoft and OpenAI declined to say anything
Since it came out at the end of November, ChatGPT has lit up the internet. It got its first million users in less than a week. Its ability to talk like a person sparked rumors that it could replace professional writers and even put Google’s search business at risk. It was started by Elon Musk and a Silicon Valley investor named Sam Altman. The organization that runs it makes money by charging developers to use its technology.
The new technology is based on OpenAI’s GPT-3 language model. It comes at the end of a year in which AI made a lot of big news. The company’s Dall-E image-generating model, which uses written instructions to make art and other images, also sparked a large debate about how AI could be used in creative fields. OpenAI is already working on a new model to replace the GPT-4 model for processing natural language.
Microsoft has put about $1 billion into OpenAI in the past. It is also trying to add ChatGPT to its Bing search engine to give it an edge over Alphabet Inc.’s search service, which is currently the most popular. The bot can answer questions in a natural and humanlike way, carry on a conversation, and answer follow-up questions, unlike a Google search, which only gives you a list of links.
Still, concerns about its accuracy—which Altman himself has said isn’t good enough for the bot to be relied on—have made people wary of using it too soon, and schools in New York City have stopped their students from using ChatGPT.