Geo News
Community curated by people like you
LatestAICryptoHealthWorld AffairsUS Politics
Bitcoin Red Team says Chinese AI models dominate security research as US models block cybersecurity queries
00

Bitcoin Red Team says Chinese AI models dominate security research as US models block cybersecurity queries

Aug 22, 2026

The volunteer-run Bitcoin Red Team, consisting of 20 to 25 security researchers, is proactively scanning the open-source Bitcoin ecosystem to preempt AI-driven cyberattacks. Member Calle reveals that the group relies heavily on Chinese AI models like Kimi K3 because strict guardrails on U.S. models block critical cybersecurity queries, even refusing to assist in fixing identified vulnerabilities. This restriction hampers defenders as AI lowers technical barriers, enabling low-skilled attackers to execute exploits end-to-end.

Bitcoin Red Team operations

  • ▪As of August 4, 2026, the Bitcoin Red Team had spent approximately $20,000 across different services to fund its security operations.
  • ▪The Bitcoin Red Team consists of approximately 20 to 25 volunteer security researchers and developers, including Cashu maintainer Calle and AnchorWatch CEO Rob Hamilton.
  • ▪The Bitcoin Red Team began forming after AnchorWatch CEO Rob Hamilton started examining Bitcoin projects following the Coldcard air-gapped wallet hack.

Chinese AI cybersecurity dominance

  • ▪Bitcoin Red Team member Calle stated that Chinese artificial intelligence models are used far more than United States models for cybersecurity research.
  • ▪Bitcoin Red Team member Calle stated that the release of the Chinese artificial intelligence model Kimi K3 has given both attackers and defenders unprecedented capabilities.

U.S. AI model restrictions

  • ▪United States artificial intelligence models often block cybersecurity-related requests and refuse to assist developers with finding or fixing software vulnerabilities due to heavy guardrails.
  • ▪Bitcoin Red Team member Calle stated that while United States frontier artificial intelligence models remain highly capable, their strict safety restrictions limit their usefulness in cybersecurity.

AI-enabled hacking threats

  • ▪Bitcoin Red Team member Calle warned that artificial intelligence has eliminated the information asymmetry that previously allowed for security through obscurity in software.
  • ▪Artificial intelligence has lowered the technical barrier to cyberattacks, allowing individuals without advanced security expertise to execute software exploits from beginning to end.

Bitcoin ecosystem vulnerability scanning

  • ▪The Bitcoin Red Team has proactively scanned almost the entire significant open-source Bitcoin ecosystem to identify and report vulnerabilities to developers.
  • ▪While the Bitcoin protocol itself remains secure, the Bitcoin Red Team focuses its security sweeps on applications, wallets, and services built around Bitcoin.

1 source

Decrypt
AI Has Made Bitcoin Software a Target—This Group Is Fighting Back - Decrypt
View source article

Featured stories

View more in China

OpenAI and Anthropic CEOs called to appear at Australian AI inquiry

Sep 27, 2026 · 2 sources

Bill Gates warns AI powerful enough to cause a billion deaths

Sep 25, 2026 · 5 sources

FTC opens investigation into OpenAI and Anthropic over consumer protection

Sep 30, 2026 · 7 sources

Trump signs voluntary AI safety accord with tech executives

Sep 29, 2026 · 14 sources

Story comments

Loading comments…

Related entities

ChinaUnited States

Related Projects

Decrypt

Topics

ChinaAI securityAI RegulationBitcoin security & risks

Featured stories

View more in China

OpenAI and Anthropic CEOs called to appear at Australian AI inquiry

Sep 27, 2026 · 2 sources

Bill Gates warns AI powerful enough to cause a billion deaths

Sep 25, 2026 · 5 sources

FTC opens investigation into OpenAI and Anthropic over consumer protection

Sep 30, 2026 · 7 sources

Trump signs voluntary AI safety accord with tech executives

Sep 29, 2026 · 14 sources