Bitcoin Red Team is forced to return to Chinese AI models due to OpenAI blocks.
The volunteer initiative Bitcoin Red Team, which uses artificial intelligence to audit the security of the Bitcoin ecosystem, has hit a serious obstacle. AnchorWatch CEO Rob Hamilton reported that on August 8 he began integrating OpenAI's cyber capabilities into his work, but access was blocked the very next morning.
Notably, the block occurred after Hamilton had completed the full KYC procedure and finished onboarding several months earlier. According to him, due to the restrictions, he was unable to continue auditing code changes and checking for other potential vulnerabilities. This creates a paradoxical situation: American AI companies are effectively cutting off defenders from the most powerful tools, while attackers remain completely free in their choice of technology.
Chinese models as an alternative
As a result, Hamilton was forced to return to using Chinese open-source AI models. In one tweet, he noted that he and another initiative participant under the handle calle are beginning a "very Chinese period of life." The Bitcoin Red Team team confirmed that OpenAI and Anthropic's cyber models do indeed deliver "stunning results," but the bulk of accumulated AI spending goes to Kimi K3 — nearly 75%. The share of Qwen 3.8 from Alibaba is also growing.
When asked about Anthropic's policy regarding trusted access to the most powerful models, Hamilton replied that no issues have arisen so far, but he expects a downgrade from the Fable model to Opus, which is essentially equivalent to restrictions.
Scope of work and conclusions
As a reminder, Bitcoin Red Team is a volunteer group using AI and manual review to analyze hundreds of open-source repositories related to Bitcoin. As of August 9, the team of 25 participants had found 1,288 critical and high-level vulnerabilities in the Bitcoin ecosystem. Cumulative AI spending has already exceeded $58,000.
This situation raises an important question about strategic dependence on American AI platforms. In July, 25 companies urged U.S. authorities not to impose restrictions on Chinese models with open weights, and this case vividly demonstrates the validity of such concerns. In conditions where the cybersecurity of critical infrastructure depends on access to advanced AI tools, artificial barriers only weaken the position of defenders, forcing them to seek alternatives outside the United States. This is not just an inconvenience — it is a strategic miscalculation that could have long-term consequences for the entire industry.