[openai-blog] Introducing ChatGPT Enterprise
OpenAI announced ChatGPT Enterprise on 28 August 2023, positioning the service as a business-focused tier with enhanced security, administrative controls, and unlimited access to GPT-4 [source]. The company stated that Enterprise customers would receive priority access to new features, higher speed limits, and advanced data analysis capabilities.
The announcement followed months of enterprise pilot testing and came as organisations raised concerns about data governance when using consumer AI tools. OpenAI emphasised that Enterprise customer data would not be used for model training, addressing a key barrier to corporate adoption.
ChatGPT Enterprise includes single sign-on, domain verification, and an admin console for user management. OpenAI stated the service operates on dedicated capacity with performance "up to two times faster" than the standard offering, though no independent benchmarks were provided at launch.
The pricing model was not disclosed in the announcement. OpenAI indicated that costs would be determined through direct sales conversations based on usage requirements and seat count.
The launch represented OpenAI's first major segmentation of ChatGPT into consumer and enterprise tiers since the product's November 2022 release. The company reported that more than 80 percent of Fortune 500 companies had employees using ChatGPT, though it did not specify whether this usage was authorised or shadow IT.
OpenAI stated that Enterprise builds on the existing ChatGPT Plus and API offerings but provides "enterprise-grade security and privacy" as distinct features. The announcement did not detail specific compliance certifications or third-party security audits completed at the time of launch.
Why this is an AI incident
Launch-archive bulk classification (10 May 2026). Source signal originates from a real AI provider, regulator, or model-comparison probe; the harm or behavioural change described would not have occurred without the AI system being deployed in the role described. Editor reviewing the archive may amend the rationale per-wire.
Counterfactual "but-for" test per the Editor's Guide.