Operator: the AI agent for browsing the internet
In each industrial revolution, a technology drives the change. In the 19th century, it was the steam engine; in the 20th, electricity and computing. Today, artificial intelligence (AI) leads this transformation. On January 23, 2025, OpenAI unveiled its latest innovation: Operator, a virtual assistant capable of browsing the internet and performing complex tasks autonomously. This technology redefines how we navigate the web, optimizing time and freeing our creativity. Moreover, its impact transforms the way businesses and professionals operate, anticipating a future where efficiency and intelligence work together seamlessly.
THE EVOLUTION OF AI
The 2010s will be remembered for the progress of one of the most advanced technologies in the world: artificial intelligence. Over the years, as more resources have been allocated to its development and it has gained greater acceptance among both consumers and businesses, it is interesting to review some of the key milestones of the past decade that have driven this progress.
>> the beginnings of artificial intelligence
Ten years ago, artificial intelligence was only used for specific tasks, such as voice recognition or image classification. It couldn't create on its own; it only analyzed what already existed. But in 2022, that changed when models like ChatGPT and DALL-E became accessible to the public. These AIs not only understood context better than before, but they could also generate all kinds of content: texts, images, even audio, without fully depending on a person.
This led companies to start using them in various areas, although at first, their integration was experimental. They improved customer service, helped optimize processes, and changed the way digital content was created. Over time, their impact has grown, transforming entire sectors and paving the way for the next major evolution of AI.
>> the arrival of AI AGENTS
Generative artificial intelligences, such as ChatGPT or Claude, started as text assistants, answering questions and creating content, but without the ability to act on their own. Some companies tried to enable them to use programs and recognize screens, but at first, they didn’t work well. Claude, for example, struggled with images and complex tasks. Microsoft, internally, has made progress in this area with agents that can manage emails and automate office tasks, improving productivity.
Now, with Operator, OpenAI has made it possible for AI not only to read and write, but also to browse the internet, interact with web pages, and perform actions within the browser without constant help. It can search for information, fill out forms, make reservations, and even make small decisions without someone telling it what to do at every moment. And not only that, but just as it did with ChatGPT, it has been designed as a product for anyone to benefit from. Currently (early 2025), Operator is exclusively available to ChatGPT Pro users in the United States. This initial research phase allows OpenAI to gather user feedback and improve Operator’s capabilities before expanding its availability. In the future, it is planned to offer access to other paying users and integrate it directly into ChatGPT.
KEY FEATURES OF OPERATOR
What makes it different? Until now, digital tools have allowed the automation of structured processes, tasks limited to following predefined rules. Operator goes one step further, addressing more fluid and contextual actions. Operator can handle everyday tasks that previously required our direct attention. From posting content on social media, to doing online shopping or booking flight tickets, this tool opens new possibilities for personal and business productivity. Unlike traditional virtual assistants, Operator has advanced capabilities that allow it to:
- Autonomous web browsing: Operator uses its own browser to explore websites, interact with graphical interfaces, and complete tasks ranging from filling out forms to finding the perfect gift for a loved one or even creating memes.
- Interaction with graphical interfaces: It is capable of interpreting and manipulating on-screen elements in a way similar to how human users do, making it unique compared to other AI systems.
- Learning and self-correction: Through use, Operator improves its skills, adapting to the specific needs of each user. If it encounters difficulties, it uses its reasoning ability to self-correct, and if it gets stuck, it returns control to the user to ensure a smooth and collaborative experience.
- Advanced security: It uses encryption protocols to ensure the integrity of business data in digital environments.
For example: I’m a shopper looking for the best-priced hotel with good reviews. Instead of manually comparing across multiple websites, Operator searches, filters, and secures the best available offer. If the booking requires payment, it asks for my approval and completes the purchase action, preventing me from losing availability at high-demand accommodations.
Tangible Benefits for Businesses and Users
- Reduction in operational costs: Operator automates repetitive tasks, optimizing resources and reducing expenses, especially in sectors with tight margins like logistics and tourism. For example, a hotel can use Operator to automatically review reviews on platforms like Google and TripAdvisor. It can browse these pages, filter negative comments, and gather the most mentioned points. This way, the staff does not have to manually go through hundreds of reviews, saving time and reducing costs.
- Increased productivity: It frees professionals from mechanical tasks, allowing them to focus on deep work and strategic decisions, driving innovation. A digital marketing team uses Operator to manage collaborations with influencers on social media. Typically, professionals have to log into platforms like Instagram or TikTok, search for brand mentions, review profiles, gather statistics, and manually enter the data into their management dashboard. Operator automates this process by browsing through networks, detecting relevant mentions, extracting engagement metrics, and organizing the information into a report. This frees specialists from repetitive tasks and allows them to focus on content strategy, influencer negotiation, and creativity in campaigns, instead of wasting time manually gathering data.
- Accuracy and consistency: It minimizes human errors in routine and repetitive processes, ensuring accurate and up-to-date data. For example, an online store uses Operator to check if the products in its catalog are still in stock on its suppliers' websites. It can browse their pages, verify availability, and update the online store without human intervention and without any human oversight, preventing customers from purchasing out-of-stock items.
- Sector adaptability: It adapts to multiple industries, enhancing customer service in tourism, managing shipments in logistics, and other key processes. A restaurant could use Operator to manage refunds on platforms like Uber Eats and Glovo, as their APIs do not allow this task to be automated. Normally, staff must manually log into each platform, find the affected order, review the refund options, and complete forms. Operator performs this process by browsing the web like a human user: it locates the order, identifies the issue, fills in the required fields, and submits the request. If the platform requires evidence, such as a photo or comment, Operator can detect it and attach the correct information. This reduces the staff's workload and speeds up issue resolution.
The Future of AI Agents: An expanding Ecosystem
In this future, we will not only see AI agents working autonomously, but also a new market of companies specialized in their development and configuration. Just as today there are companies that create custom software, in the coming years, businesses will emerge dedicated to designing customized AI agents for different industries. Companies will not only adopt AI agents, but they will also seek solutions tailored to their specific needs. Some will require AI agents that manage their finances, others will need advanced assistants for customer service, and many will want to integrate these systems into their internal processes efficiently.
The rapid integration of AI agents into the workforce is giving rise to a new profession: the AI agent manager. These professionals will be responsible for overseeing, optimizing, and coordinating the interaction between AI agents and human employees, ensuring efficient and harmonious collaboration.
According to an article from Forbes, in the next three years, agent-based AI will impact all areas of business, transforming interactions between humans and machines. Leading companies are already exploring how to integrate and manage these agents within their teams. The emergence of AI agent managers highlights the need for new skills in the workforce, focusing on supervision and collaboration with advanced technologies.
Conclusion
The arrival of AI agents marks a profound shift in the way we work, moving from simple task automation to systems capable of executing actions autonomously and adaptively. OpenAI is already collaborating with companies such as DoorDash, OpenTable, Uber, and StubHub to explore use cases in everyday life and integrate them as predefined actions in Operator, facilitating their adoption across different sectors.
Currently, Operator is available in a preliminary research version, which implies limitations in its functionality and processing capacity. The organization led by Sam Altman promises that its product will evolve based on feedback received from users, adapting to the real needs of the market.
In the future, OpenAI plans to improve the precision and efficiency of Operator in long and complex workflows, enabling the automation of more sophisticated processes within the browser. Additionally, it will launch the AI model powering Operator in the API, opening up the possibility for developers to build their own customized agents based on this technology, further expanding its applications in the business world.
