OpenAI's Codex now runs on Windows 11 with a "Computer Use" feature that allows the AI to independently control programs, test applications, and identify bugs without human intervention.
OpenAI has expanded Codex capabilities with a new "Computer Use" feature for Windows 11. The AI can now operate your PC autonomously, performing tasks like debugging code, testing software, and managing applications independently.
The system leverages ChatGPT's mobile app to enable remote task management. Users can initiate and monitor operations from their phone while away from their desk, giving developers flexibility in how they deploy automated testing and debugging workflows.
Codex, built on GPT-4, was designed to understand and generate code. This Windows integration represents a significant expansion of its practical applications. Rather than simply suggesting code fixes, the AI can now execute actions directly within the operating system and its applications.
The autonomous capabilities address a common development bottleneck: repetitive testing and bug identification. By automating these tasks, teams could allocate engineering resources to higher-level problem-solving and feature development.
The mobile integration allows developers to queue up testing sessions before leaving the office or initiate debugging runs remotely. The ChatGPT app serves as the control interface, displaying task progress and results.
This development reflects a broader trend of AI systems moving beyond text generation into direct software operation. As AI models become more capable, their integration into existing workflows continues to deepen.
The feature availability on Windows 11 suggests OpenAI is prioritizing enterprise and developer adoption. Windows remains dominant in corporate development environments, making it a logical platform for deployment.
Details on system requirements, pricing, or availability timelines for general users remain limited. OpenAI has been selective about rolling out Computer Use capabilities across its product lineup.
The announcement underscores ongoing competition in the AI development tools space, where companies like GitHub Copilot and others continue introducing automation features aimed at streamlining software development cycles.
At a Singapore defense forum, panelists identified artificial intelligence as a greater strategic threat than nuclear weapons, citing concerns that AI could compress decision-making timelines to dangerous levels.
TikTok sellers are using AI-generated personas to market low-cost products, with some accounts depicting fake Black women in emotional sales pitches. The accounts blur the line between authentic influencer marketing and deceptive commerce.
AI startup Shift is cleaning homes at no cost to homeowners while collecting data to train autonomous cleaning robots. The company uses human cleaners to generate training footage for future robotic systems.
A large-scale study of 26 million responses shows that training language models to be helpful chatbots simultaneously weakens their ability to simulate human behavior. The problem intensifies with each new model generation.