OpenAI’s GPT-5.4 introduces computer-use capabilities, a one-million-token context window, and improved reasoning—signaling a shift toward AI systems that can function as digital workers.
 |
| OpenAI’s GPT-5.4 expands AI capabilities with automation and large-scale data analysis, highlighting a new phase in the global race to build AI-powered digital workers. Image: CH |
Tech Desk — March 6, 2026:
The launch of GPT-5.4 by OpenAI marks a pivotal development in the evolution of artificial intelligence, signaling a shift toward systems capable of performing complex digital tasks rather than merely generating information.
The latest model introduces several major upgrades, including improved reasoning, enhanced coding capabilities and automated computer-use functions. The company says the system integrates multiple technological advances into a unified platform available in variants such as GPT-5.4 Thinking and GPT-5.4 Pro.
One of the most significant improvements is the model’s one-million-token context window, enabling it to process extremely large datasets. This expanded capacity allows the AI to analyse entire software codebases, lengthy academic research papers and extensive document archives without splitting them into smaller segments.
Perhaps the most notable change is the model’s built-in computer-use capability, allowing AI agents to interact directly with software applications. By interpreting screenshots and issuing mouse clicks and keyboard commands, GPT-5.4 can navigate websites, control programs and automate complex workflows.
This development represents a fundamental shift in how AI systems function. Earlier versions of ChatGPT focused mainly on answering questions and generating text. Later models such as GPT-4 expanded into advanced writing, summarisation and coding.
With GPT-5.4, however, the technology moves beyond generating suggestions to actually executing tasks on computers, a step many experts see as critical for integrating AI into everyday workplace operations.
OpenAI says the model is designed primarily for professional and enterprise environments. Its improved reasoning and long-task management allow it to handle complex workflows in areas such as financial analysis, legal research and software development.
For example, the system can analyse financial data in spreadsheets using programs such as Microsoft Excel, generate dashboards, create reports from raw datasets and process large volumes of contractual or legal documents.
In software engineering environments, GPT-5.4 can generate large codebases, identify and fix bugs, run automated tests and control web browsers through automation tools.
Developers can also use the model to build AI agents capable of planning tasks, executing them and adjusting strategies when obstacles appear—an ability that moves AI closer to functioning as an autonomous digital assistant.
The company also says GPT-5.4 reduces the likelihood of “hallucinations,” a term used to describe when AI models generate inaccurate or fabricated information. According to OpenAI, the new model is about 33 percent less likely to produce false information compared with earlier versions.
While not eliminating the issue entirely, the improvement is important for professional users who rely on AI for high-stakes tasks such as legal analysis, research and financial modelling.
The release comes amid intensifying competition in the global artificial intelligence sector. Rival firm Anthropic, led by Dario Amodei, recently introduced its own advanced models, including Claude Opus 4.6 and Claude Sonnet 4.6.
These systems aim to deliver faster and more efficient performance for enterprise tasks, reflecting the industry’s broader push to develop AI platforms capable of serving as practical digital workers.
The rapid evolution of models such as GPT-5.4 highlights a major transition in artificial intelligence. Early chatbots focused on conversation and information retrieval. Today’s models increasingly combine reasoning, automation and software interaction to perform real tasks across digital environments.
As companies continue to invest in increasingly capable AI systems, experts say the next stage of innovation may center on AI agents that collaborate with humans or operate independently within digital workplaces.
For businesses and professionals, the implications could be significant: a future where AI systems not only assist with information but actively help run workflows across industries.