OpenAI Launches Operator: A New AI Tool for Automating Web-Based Tasks

On January 23, 2025, OpenAI introduced Operator, a new AI-powered tool available as a research preview exclusively for Pro users in the United States. Operator uses its own browser to independently perform user-defined tasks, aiming to simplify repetitive activities such as filling out forms, ordering groceries, or creating digital content.

OpenAI Launches Operator: A New AI Tool for Automating Web-Based Tasks

On January 23, 2025, OpenAI introduced Operator, a new AI-powered tool available as a research preview exclusively for Pro users in the United States. Operator uses its own browser to independently perform user-defined tasks, aiming to simplify repetitive activities such as filling out forms, ordering groceries, or creating digital content.

At the core of Operator is the Computer-Using Agent (CUA) model. This technology combines the visual capabilities of GPT-4o with advanced reasoning skills built through reinforcement learning. CUA enables interaction with graphical user interfaces (GUI), allowing it to click buttons, fill in text fields, and navigate menus on a browser. This functionality equips Operator to mimic the tasks typically performed by a human user. Additionally, it has self-correcting capabilities to address errors and can return control to the user when needed.

How Operator Works

Operator is designed to execute web-based tasks with a user-like approach. It “sees” web pages through screenshots and interacts using mouse and keyboard commands. Operator has already set benchmarks in browser task execution with high performance on platforms like WebArena and WebVoyager. Users can assign tasks and let Operator handle the rest. When sensitive actions, such as entering payment details, are required, Operator asks the user to take over to ensure a secure experience.

The tool is particularly effective for personalized tasks. For instance, users can save frequently purchased items for quick grocery orders or set preferences for booking travel. Operator also supports multitasking, allowing users to simultaneously order a personalized mug from Etsy while reserving a campsite on Hipcamp.

Collaborations and Public Sector Potential

Operator is being developed in collaboration with companies such as Instacart, DoorDash, OpenTable, Priceline, StubHub, Thumbtack, and Uber to meet real-world needs. Additionally, the tool holds potential for the public sector, aiming to streamline processes like enrolling in city services. For instance, the City of Stockton is working with OpenAI to enhance accessibility for its residents.

According to Jamil Niazi, Director of Information Technology for the City of Stockton, “As we learn more about Operator during its research preview, we’ll be better equipped to identify ways that AI can make civic engagement even easier for our residents.”

Privacy and Security Features

Operator prioritizes security with three layers of safeguards to keep users in control:

  1. User Control: For sensitive tasks like entering passwords or payment information, Operator requests the user to take over. It neither stores nor screenshots this data.
  2. User Confirmation: Critical actions, such as submitting an order or sending an email, require explicit user approval.
  3. Task Restrictions: Operator is trained to avoid high-risk tasks, such as financial transactions or decisions involving significant consequences.

Operator also offers robust privacy options. Users can opt out of contributing to model training by disabling the “Improve the Model for Everyone” feature. Additionally, browsing data can be deleted with a single click, and all accounts can be logged out simultaneously through the privacy settings.

Development and Limitations

As an early-stage research preview, Operator has some limitations. It may face challenges with complex interfaces, such as creating slideshows or managing calendars. Feedback from users will play a crucial role in improving its reliability and functionality.

Daniel Danker, Chief Product Officer at Instacart, remarked, “OpenAI's Operator is a technological breakthrough that makes processes like ordering groceries incredibly easy.”

Future plans for Operator include expanding its capabilities to handle more complex workflows and introducing the CUA model through APIs for developers who wish to build their own browser-based AI tools. OpenAI also intends to make Operator accessible to Plus, Team, and Enterprise users, integrating it into ChatGPT for seamless task execution.

By launching Operator, OpenAI aims to simplify digital workflows and help users save time on repetitive tasks. As the tool evolves through user feedback, it has the potential to become a key player in enhancing digital accessibility and efficiency for both individuals and businesses.

Contributors to the Operator Application

  • CEO: Sam Altman
  • Research Leads: Hyeonwoo Noh, Reiichiro Nakano
  • Model Safety and Readiness Lead: Casey Chu
  • Infrastructure Lead: Aaron Schlesinger
  • Research Infrastructure Lead: David Medina
  • Product Director: Daniel Danker

Core Research Contributors: Casey Chu, David Medina, Hyeonwoo Noh, Noah Jorgensen, Reiichiro Nakano, Sarah Yoo

Core Team: Andrew Howell, Aaron Schlesinger, Baishen Xu, Ben Newhouse, Bobby Stocker, Devashish Tyagi, Dibyo Majumdar, Eugenio Panero, Fereshte Khani, Geoffrey Iyer, Jiahui Yu, Nick Fiacco, Patrick Goethe, Sam Jau, Shunyu Yao, Stephan Casas, Yash Kumar, Yilong Qin

XFN Contributors: Abby Fanlo Susk, Aleah Houze, Alex Beutel, Alexander Prokofiev, Andrea Vallone, Andrea Chan, Christina Lim, Derek Chen, Duke Kim, Grace Zhao, Heather Whitney, Houda Nait El Barj, Jake Brill, Jeremy Fine, Joe Fireman, Kelly Stirman, Lauren Yang, Lindsay McCallum, Leo Liu, Mike Starr, Minnia Feng, Mostafa Rohaninejad, Oleg Boiko, Owen Campbell-Moore, Paul Ashbourne, Stephen Imm, Taylor Gordon, Tina Sriskandarajah, Winston Howes

Additional Contributors: Adam Brandon, Adam Koppel, Adele Li, Ahmed El-Kishky, Akila Welihinda, Alex Karpenko, Alex Nawar, Alex Tachard Passos, Amelia Liu, Andrei Gheorghe, Andrew Duberstein, Andrey Mishchenko, Angela Baek, Ankush Agarwal, Anting Shen, Antoni Baum, Ari Seff, Ashley Tyra, Behrooz Ghorbani, Bo Xu, Brandon McKinzie, Bryan Brandow, Carolina Paz, Cary Hudson, Chak Li, Chelsea Voss, Chen Shen, Chris Koch, Christian Gibson, Christina Kim, Christine McLeavey, Claudia Fischer, Cory Decareaux, Daniel Jacobowitz, Daniel Wolf, David Kjelkerud, David Li, Ehsan Asdar, Elaine Kim, Emilee Goo, Eric Antonow, Eric Hunter, Eric Wallace, Felipe Torres, Fotis Chantzis, Freddie Sulit, Giambattista Parascandolo, Hadi Salman, Haiming Bao, Haoyu Wang, Henry Aspegren, Hyung Won Chung, Ian O’Connell, Ian Sohl, Isabella Fulford, Jake McNeil, James Donovan, Jamie Kiros, Jason Ai, Jason Fedor, Jason Wei, Jay Dixit, Jeffrey Han, Jeffrey Sabin-Matsumoto, Jennifer Griffith-Delgado, Jeramy Han, Jeremiah Currier, Ji Lin, Jiajia Han, Jiaming Zhang, Jiayi Weng, Jieqi Yu, Joanne Jang, Joyce Ruffell, Kai Chen, Kai Xiao, Kevin Button, Kevin King, Kevin Liu, Kristian Georgiev, Kyle Miller, Lama Ahmad, Laurance Fauconnet, Leonard Bogdonoff, Long Ouyang, Louis Feuvrier, Madelaine Boyd, Mamie Rheingold, Matt Jones, Michael Sharman, Miles Wang, Mingxuan Wang, Nick Cooper, Niko Felix, Nikunj Handa, Noel Bundick, Pedro Aguilar, Peter Faiman, Peter Hoeschele, Pranav Deshpande, Raul Puri, Raz Gaon, Reid Gustin, Robin Brown, Rob Honsby, Saachi Jain, Sandhini Agarwal, Scott Ethersmith, Scott Lessans, Shauna O’Brien, Spencer Papay, Steve Coffey, Tal Stramer, Tao Wang, Teddy Lee, Tejal Patwardhan, Thomas Degry, Tomo Hiratsuka, Troy Peterson, Wenda Zhou, William Butler, Wyatt Thompson, Yao Zhou, Yaodong Yu, Yi Cheng, Yinghai Lu, Younghoon Kim, Yu-Ann Wang Madan, Yushi Wang, Zhiqing Sun

Leadership: Anna Makanju, Hannah Wong, Jerry Tworek, Liam Fedus, Mark Chen, Peter Welinder, Sam Altman