Artificial Intelligence

Preferred on Google

ChatGPT Gains Computer Control Capabilities

OpenAI introduces new ChatGPT capabilities allowing it to control websites and desktop applications, enhancing productivity through browser integration and computer control.

Jul 16 at 7:02 PM7 min read

Man sitting at a desk with a laptop, with text overlays indicating 'Browser and Computer Use with the new ChatGPT app'. — OpenAI Youtube

Visual TL;DR. ChatGPT new capabilities enables Control websites. ChatGPT new capabilities enables Control desktop apps. Control websites includes Seamless browser integration. Control websites example Real-world tasks. Control desktop apps example Real-world tasks. Control websites leads to Enhance productivity. Control desktop apps leads to Enhance productivity. ChatGPT new capabilities explained by Developer Experience.

ChatGPT new capabilities: OpenAI expands ChatGPT to control websites and desktop applications directly
Control websites: in-app browser can scan websites to gather information and compile reports
Seamless browser integration: annotation tool allows targeted feedback or questions about specific web pages
Control desktop apps: ChatGPT can now complete tasks on your computer, from OpenAI YouTube
Enhance productivity: transforming ChatGPT from conversational tool into proactive assistant executing complex workflows
Real-world tasks: AI performs tasks like analyzing community forum posts to identify user frustrations
Developer Experience: Dominik Kundel explains new ChatGPT app integrates with tools users already employ

Visual TL;DRQuickExplainDeeper

OpenAI has unveiled a significant expansion of ChatGPT's capabilities, allowing the AI to perform real-world tasks by controlling websites and applications directly. This update aims to transform ChatGPT from a conversational tool into a proactive assistant capable of executing complex workflows.

Browser and Computer Control

Dominik Kundel, Developer Experience at OpenAI, explains that the new ChatGPT app integrates with the tools users already employ. A key feature is the in-app browser, which can scan websites to gather information and even compile reports. For instance, ChatGPT can analyze community forum posts to identify user frustrations over a specific period.

The full discussion can be found on OpenAI Youtube's YouTube channel.

ChatGPT can now complete tasks on your computer - OpenAI Youtube — ChatGPT can now complete tasks on your computer, from OpenAI Youtube

The in-app browser also includes an annotation tool. This allows users to provide targeted feedback or ask questions about specific web pages without needing to craft lengthy, detailed instructions. This granular control enhances the AI's ability to understand and act on user prompts within a web context.

Seamless Browser Integration

For users who prefer to maintain their existing browser environment, ChatGPT now offers a Chrome extension. This extension enables ChatGPT to access tabs that are already open or create new ones in the background. This passive operation ensures that users can continue their work without interruption while ChatGPT performs tasks like filing expense reports or managing long data entry processes.

Desktop Application Control

Beyond web browsing, ChatGPT can now control applications installed on a user's desktop. This 'Computer Use' functionality extends to both Windows and macOS operating systems. On macOS, ChatGPT features its own cursor, allowing it to operate in the background while the user engages with other applications. This capability is demonstrated by updating notes with to-do lists, showcasing ChatGPT's ability to interact with native apps like Apple Notes.

This advancement signifies a shift towards AI agents that can more deeply integrate into users' daily digital lives. By connecting directly to the tools and applications people use regularly, the new ChatGPT aims to boost productivity and streamline complex tasks.

© 2026 StartupHub.ai. All rights reserved. Do not enter, scrape, copy, reproduce, or republish this article in whole or in part. Use as input to AI training, fine-tuning, retrieval-augmented generation, or any machine-learning system is prohibited without written license. Substantially-similar derivative works will be pursued to the fullest extent of applicable copyright, database, and computer-misuse laws. See our terms.

#OpenAI #ChatGPT #Artificial Intelligence