Artificial intelligence is seeing a rise in "agent wave". OpenAI will soon launch an AI agent tool called Operator.

According to two sources familiar with the matter, OpenAI is preparing to launch a new type of artificial intelligence agent codenamed "Operator" that can use computers to perform human actions such as coding or booking travel. One of the sources said that during a staff meeting on Wednesday, OpenAI leadership announced plans to release the tool as a research preview in January next year and make it available to developers through the company's application programming interface. OpenAI did not immediately respond to requests for comment. The release of this plan is part of a broader industry trend towards advancing artificial intelligence agent software, which can complete multi-step tasks with minimal supervision for users. Anthropic has launched a similar agent that can process events happening on a user's computer in real-time and take actions on behalf of the user. OpenAI supporter Microsoft Corporation (MSFT.US) also recently introduced a set of agent tools designed to send emails and manage records for employees. Alphabet Inc. Class C (GOOGL.US) is reportedly preparing to release an artificial intelligence agent, according to The Information. According to three sources familiar with the matter, OpenAI has been conducting several research projects related to agents. One source said that the closest to completion is a general-purpose tool for performing tasks in web browsers. OpenAI CEO Sam Altman hinted at the shift towards agents during a "Ask Me Anything" session on Reddit last month. Altman wrote, "We will have better and better models," "But I think the next major breakthrough will be agents." As OpenAI and its competitors invest heavily in developing more advanced artificial intelligence models, the returns on their investments are diminishing.