Recently, reports emerged indicating that Google has started blocking paying subscribers of its AI service who utilize the third-party tool OpenClaw to access its Gemini models. These restrictions primarily impact users of the Google AI Ultra subscription, which costs as much as $249.99 per month. Many users have voiced their frustration after facing sudden access blocks without any prior notice.
Peter Steinberger, the creator of the open-source AI agent OpenClaw, has also encountered these blocks. He remarked, “It seems quite severe on Google’s part. Be cautious if you’re using Antigravity. I’m considering withdrawing my support.” He contrasted Google’s approach with that of Anthropic, noting that they handle issues more amicably: “Anthropic contacts me with friendly assistance, while Google’s method seems to be outright blocking.”
OpenClaw is designed to allow users to utilize various AI models; although Steinberger has transitioned to using OpenAI, the AI agent remains independent of any single large language model (LLM) provider.
Background of the Blocks
Google reportedly justifies these actions on multiple grounds. Central to their reasoning is the claim that OpenClaw’s use violates their terms of service, specifically regarding the proper use of Google Antigravity OAuth tokens, which are intended for official operations solely on the Google platform.
Furthermore, it is alleged that OpenClaw generates continuous automated calls that Google’s systems identify as abnormal usage patterns. Such automation could be perceived as malicious activity, which could negatively impact service quality, according to Google’s rationale.
Security Concerns and Industry Trends
Security issues have also been raised in connection with OpenClaw’s proliferation. The rapid uptake of the tool has sparked discussions around potential security vulnerabilities, data leaks, and the unpredictability associated with its use. Reports indicate that some instances of OpenClaw may have been exposed, raising concerns about information theft, especially in enterprise settings.
Notably, Google’s actions align with a broader industry trend. Anthropic, another prominent AI provider, recently amended its terms of service to prohibit the utilization of OAuth tokens in third-party tools, citing unusual traffic patterns and debugging difficulties as the reasons behind this policy change.
Impact on Affected Users
The fallout for users who have been blocked often extends beyond just losing access to Gemini 2.5 Pro. In some instances, other linked services, such as Gmail and Google Workspace, have also faced restrictions. This situation is particularly troubling as users continue to be charged for their subscriptions while their access remains curtailed.
Frustration has been voiced in various forums, including a Google-owned forum, as well as on Hacker News and Reddit. Users have expressed disappointment with Google’s lack of communication, the abrupt nature of the blocks, and challenges in obtaining support, contributing to widespread dissatisfaction.
Google’s Response and Alternatives
In response, Google has stated that some users may not have received notifications about the violation of the terms of service. The company is reportedly exploring options to restore access for these individuals, although current resources are limited.
As an alternative, Google recommends utilizing official API keys through Google AI Studio or Google Cloud API keys. This would offer paid, yet compliant, traceable, and scalable access that adheres to the stipulated terms of service.
Open Questions
The total number of users affected by these blocks and the fate of their already-paid subscription fees remains uncertain. Additionally, the criteria Google employs to differentiate between intentional and unintentional violations have not been publicly disclosed.
This development raises significant questions regarding the management of third-party tools within the AI industry and may set a precedent for other providers to follow.
