OpenAI allows some users to try a new ChatGPT feature that uses its artificial intelligence to use a web browser to book travel, shop, search for deals, and perform many other tasks online.
The new tool, called Operator, is an AI agent: it relies on an AI model trained on both text and images to interpret commands and understand how to use a web browser to execute them. OpenAI claims to have the potential to automate many everyday tasks and everyday errands.
OpenAI operator tracks rival versions of both Google and Anthropic, which have those demonstrated able to use the web. AI agents are widely considered the next step in evolution for AI that tracks chatbots, and many companies have jumped on the hype bandwagon by touting them. In most cases, their capabilities are very limited and simply use a language model to automate tasks normally performed with traditional software.
“AI is evolving from a tool that can answer your questions to a tool that can also take action in the world, executing complex, multi-step workflows,” says Peter Welinder, vice president of product at OpenAI . “We will see a huge impact on people’s productivity, but also on the quality of work they are able to do. »
OpenAI admits that giving ChatGPT access to a web browser introduces new risks and says the operator can sometimes misbehave. It claims to have implemented various new protection measures and plans to gradually expand the operator’s capabilities.
Welinder and Yash Kumar, product and engineering managers for OpenAI’s Computer Using Agent, say the goal is to learn from how people use the tool. They acknowledge that the tool could make unwanted reservations or purchases, but add that a lot of work has been done to ensure that it asks before doing anything risky. “He will come back to me and ask for confirmations before taking any action that might be irreversible,” says Kumar.
OpenAI also released a new “system map” today outlining issues that could arise with Operator. These include the possibility that it will misunderstand commands or deviate from what a user requests; be misused by users; or be the target of cybercriminals.
“It also poses an incredible number of security challenges,” says Kumar. “Because your attack vector area and your risk vector area increase quite significantly.”
Operator will initially be available as a “search preview” to ChatGPT users with a Pro account, which costs $200 per month. The company says it plans to expand access while rolling out the tool slowly because it will inevitably make mistakes along the way.
In several demonstrations, Operator showed the potential for AI to take a more active role as a web assistant. The tool includes a remote web browser and a chat window for communicating with a user.
At WIRED’s request, the operator was asked to book an Amtrak train trip from New Haven, Connecticut, to Washington, DC. He went to the correct website and correctly entered the information needed to view the schedule, then requested further instructions. If a user were logged into the Amtrak website or a browser profile with stored credit card information, the operator would be able to reserve a ticket, although it is designed to first request the authorisation.
Kumar asked Operator to reserve a table at Beretta, a restaurant in San Francisco. The program went to the OpenTable website, found the right restaurant, and searched for availability before asking what to do next. OpenAI claims to have partnered with a number of popular sites, including OpenTable, to ensure Operator works well on them.
The new tool is based on OpenAI’s GPT-4o AI model, which can perceive a browser and web page and converse in typed text. The tool incorporates additional training designed to help them understand how to perform online tasks. OpenAI will also make its IT User Agent available through its API.
