Microsoft-backed OpenAI releases GPT-4, calls latest AI model “multimodal”

OpenAI, the creator of chatbot sensation ChatGPT, said on Tuesday it was beginning to release a powerful artificial intelligence model known as GPT-4, paving the way for even more human-like technology to proliferate.

The Microsoft-funded startup said in a blog post that its latest technology is “multimodal,” meaning images as well as text prompts can prompt it to generate content. Text input capabilities will be available to ChatGPT Plus subscribers and software developers first, with a waiting list, while image input capability remains a preview for its research.

The highly anticipated launch shows how office workers can turn to ever-improving AI for even more tasks, as well as how tech companies are competing to win business with such advancements. Alphabet’s Google on Tuesday announced a “magic wand” for its collaboration software that can author virtually any document, days before Microsoft showed off AI for its rival word processor that’s likely powered by OpenAI.

The startup’s latest technology was in some cases a big improvement over its previous version known as GPT-3.5, he said. In a simulation of the bar exam required of U.S. law school graduates before professional practice, the new model scored about 10% of top candidates, versus the old model scoring in the bottom 10%, OpenAI said. .

While the two versions may sound similar in casual conversation, “the difference appears when the task complexity reaches a sufficient threshold,” OpenAI said, noting that “GPT-4 is more reliable, creative, and able to handle instructions much more nuanced. .”

© Thomson Reuters 2023