OpenAI has opened ChatGPT subscription accounts to work with OpenClaw, the record-setting open-source AI agent framework with over 3 million users, allowing autonomous AI agents powered by GPT-5.4 for $23 per month. This move contrasts sharply with Anthropic’s recent decision to block their Claude model subscriptions from similar third-party agent use, signaling divergent strategies in the rapidly evolving AI ecosystem.
- OpenClaw rapidly grew to 346,000 GitHub stars and 3.2M users since Nov 2025
- OpenAI enables ChatGPT Plus users to run AI agents for $23/month via OpenClaw
- Anthropic blocks Claude subscriptions on OpenClaw citing cost concerns
What happened
OpenAI has integrated its ChatGPT subscription service with OpenClaw, an open-source AI agent framework that has quickly become the most popular open-source project in history. OpenClaw allows users to run autonomous AI agents locally, connecting to a variety of language models and messaging platforms. Subscribers to ChatGPT Plus can now authenticate using their accounts and access GPT-5.4 capabilities through OpenClaw for $23 per month.
This development comes after the framework’s remarkable growth since its launch in November 2025, reaching over 346,000 GitHub stars and 3.2 million users. The OpenClaw framework enables seamless, continuous background operation of AI agents that perform complex tasks such as managing calendars, writing code, and orchestrating workflows, all while keeping data on users’ devices.
Why it matters
OpenAI’s decision to support ChatGPT subscriptions on OpenClaw signals a strategic emphasis on broad distribution and open infrastructure, potentially fueling adoption and innovation in autonomous AI agents. By acting as the authentication and billing layer, OpenAI solidifies its role behind the scenes of this soaring open-source success, without building the agent itself.
In contrast, Anthropic has recently restricted access to its Claude AI model for use with OpenClaw by blocking subscription plans, citing economic unsustainability due to the heavy compute load autonomous agents require. This divergence highlights a competitive split in the AI industry where OpenAI pursues an open, agent-driven future, while Anthropic protects profitability by limiting high-usage subscription scenarios.
What to watch next
Observers should monitor the evolving dynamics between open AI agent frameworks like OpenClaw and commercial AI providers’ subscription models. OpenAI’s integration could accelerate mainstream adoption of autonomous agents, prompting other companies to reconsider how they balance accessibility and cost management.
Additionally, the success of OpenClaw’s open infrastructure model could influence venture capital and enterprise strategies, as investors and firms decide whether to back open layer technologies or maintain proprietary controls. Further announcements about OpenClaw’s foundation status and continued OpenAI support will also be key signals for the AI ecosystem’s future direction.