Nearly four months after OpenAI surprised the tech industry with ChatGPT, the company is releasing its next-gen version of the technology that powers the viral chatbot tool.
In a blog post on Tuesday, OpenAI unveiled GPT-4, which the company says is able to perform well on a range of standardized tests and is also less likely to “get out of the way” with their responses, as some users have already experienced.
OpenAI said the updated technology passed a mock law school bar exam with a score of about 10% of top candidates; however, the previous version, GPT-3.5, scored around the bottom 10%. GPT-4 can also read, parse or generate up to 25,000 words of text and write code in all major programming languages, according to the company.
OpenAI described the update as the “final step” for the company. While still “less capable” than humans in many real-world scenarios, it exhibits “human-level performance across a variety of professional and academic criteria,” according to the company.
GPT-4 is the latest version of OpenAI’s large language model, which is trained on large amounts of online data to generate compelling responses to user prompts. The updated version, which is now available through a waitlist, is already making its way into some third-party products, including Microsoft’s AI-powered Bing.
“We’re pleased to confirm that the new Bing runs on GPT-4, which we’ve customized for search,” Microsoft said Tuesday. “If you’ve used the new Bing preview anytime in the past five weeks, you’ve already experienced an early version of this powerful model.”
Although ChatGPT has impressed many users with its ability to generate original essays, stories and song lyrics in response to user prompts since its launch in November 2022, it has also raised some concerns. AI chatbots, including tools from Microsoft and Google, have been called out in recent weeks for being emotionally responsive, making factual errors and engaging in genuine “hallucinations,” as the industry calls it. industry.
GPT-4 has similar limitations as previous GPT models. “It’s still flawed, still limited, and it still looks more impressive on first use than after spending more time with it,” OpenAI CEO Sam Altman wrote in a series of tweets Tuesday announcing the update.
But there are noticeable improvements, he said. “It’s more creative than previous models, it hallucinates a lot less, and it’s less biased,” he said. writing.
Still, the company said, “Extreme care should be taken when using language model outputs, especially in high-stakes contexts.”
The news comes two weeks after OpenAI announced that it was opening up access to its ChatGPT tool to third-party companies, paving the way for the chatbot to be integrated into many apps and services.
Instacart, Snap and the Quizlet tutoring app are among the first partners to experiment with the tool. In January, Microsoft confirmed it was making a “multi-billion dollar” investment in OpenAI and has since rolled out the technology to some of its products, including its Bing search engine.