OpenAI has announced the release of GPT-5.5, positioning it as its most capable and intuitive model to date and a major step toward redefining how users complete complex tasks on computers.

The new model is designed to handle multi-step, ambiguous workflows with minimal guidance, marking a shift toward more “agentic” AI systems that can independently plan, execute, and refine tasks. GPT-5.5 is capable of writing and debugging code, conducting online research, analysing data, generating documents and spreadsheets, and navigating across software tools to complete end-to-end assignments.

Unlike earlier models that required users to break down tasks into structured prompts, GPT-5.5 can process loosely defined instructions and autonomously determine the steps needed to reach a desired outcome. This allows users to delegate complex, multi-part work while the model manages execution, validation, and iteration.

A key advancement lies in GPT-5.5’s improved reasoning across extended contexts, particularly in areas such as agentic coding, computer use, knowledge work, and early-stage scientific research. These domains benefit from the model’s ability to sustain logical consistency while taking action over time.

Despite its higher intelligence, GPT-5.5 maintains the same per-token latency as its predecessor, GPT-5.4, addressing a common trade-off between capability and speed in large AI systems. Additionally, the model demonstrates greater efficiency by using fewer tokens to complete coding tasks, reducing computational overhead while improving output quality.

OpenAI emphasised that GPT-5.5 is being released with its most comprehensive safety framework to date. The model has undergone extensive evaluation across internal safety benchmarks and preparedness protocols, including targeted testing in sensitive areas such as cybersecurity and biological applications.

The company also collaborated with internal and external red teams and incorporated feedback from nearly 200 early-access partners to refine real-world performance and mitigate risks before launch.

GPT-5.5 is now rolling out to Plus, Pro, Business, and Enterprise users within ChatGPT and Codex, while a higher-performance variant, GPT-5.5 Pro, is being introduced for Pro, Business, and Enterprise tiers.

OpenAI noted that API availability will follow soon, with additional safeguards being implemented to support secure and scalable deployment for enterprise and developer use cases.

The release underscores a broader industry shift toward AI systems that function less as passive tools and more as active collaborators. By combining advanced reasoning with the ability to act across digital environments, GPT-5.5 signals a move toward more autonomous, outcome-driven computing.

As organisations increasingly integrate AI into workflows, models like GPT-5.5 are expected to play a central role in enhancing productivity, accelerating research, and reshaping how complex work gets done.