With the addition of a function to automate chores like vacation planning, form completion, restaurant bookings, and grocery ordering, OpenAI is advancing its ChatGPT chatbot.
Operator is the name of the gadget that was shown on Thursday. According to OpenAI, it’s an agent that can navigate the web and complete activities for you. It’s also been trained to interact with the buttons, menus, and text fields that people use on a regular basis.
In order to further customize the activities it performs, it can also ask follow-up inquiries, like for other website login credentials. The screen can be taken over by users at any moment.
One of our first agents—AIs that can work for you on their own—is Operator, according to a blog post published by OpenAI on Thursday. It will carry out the task you set it.
Operator is currently only accessible to ChatGPT Pro users. Access to it is available at Operator.ChatGPT.com. According to OpenAI, it eventually intends to integrate Operator into ChatGPT and extend its reach to Plus, Team, and Enterprise users. The business added that it now struggles with a few activities, like making presentations and maintaining calendars.
According to OpenAI, which is supported by Microsoft, users can choose not to have some of the company’s training data collected by disabling ChatGPT’s “Improve the model for everyone” option, which prevents Operator data from being used to train the company’s models. In the privacy area, the corporation also stated that customers can log out of all websites and remove all browsing data with a single click.
An earlier release from Anthropic, the Amazon-backed AI business that created the Claude chatbot and was started by former OpenAI research executives, directly competes with Operator.
Anthropic unveiled Computer Use in October, enabling its AI agents to do intricate activities using computers in a manner similar to that of humans. According to Anthropic, it can use any program and browse the internet in real time, as well as read what is displayed on a computer screen, choose buttons, type text, navigate websites, and carry out activities.
In an interview at the time, Anthropic’s chief science officer, Jared Kaplan, told CNBC that the tool could use computers in a manner that was almost identical to our own. According to him, it can complete jobs that require tens or even hundreds of steps.
Within ten years, the generative AI market—which comprises OpenAI, Anthropic, Google, Amazon, Microsoft, and Meta—is expected to generate over $1 trillion in revenue.
A source familiar with the matter told CNBC this week that Google just consented to a new investment of over $1 billion in Anthropic. According to a CNBC story earlier this month, Anthropic is in final negotiations to raise $2 billion in a fundraising round at a $60 billion valuation determined by Lightspeed Venture Partners.
A possible future of artificial general intelligence is being pushed toward by OpenAI. AI that meets or exceeds human intellect on a variety of activities is referred to by the ill-defined standard known as artificial general intelligence (AGI).
In a Thursday interview with CNBC, Alexandr Wang, CEO of Scale AI, whose company supplies training data to major AI companies, stated that he defines artificial general intelligence (AGI) as strong AI systems that can operate a computer in the same way that you or I could. According to him, it will probably take two to four years for technology to advance to that point.
More from CNBC:
-
Morgan Stanley CEO Ted Pick says bank will work with U.S. regulators on crypto
-
Shein exec says tariffs shouldn t impact its cheap clothes
-
New Christie s CEO sees opportunity in younger collectors and crypto