According to @timnitGebru, Anthropic, a self-described AI safety company, allegedly leaked its entire source code, raising red flags for governments integrating Claude into critical infrastructure; as reported by The Guardian, Anthropic’s Claude code was exposed, heightening concerns over model supply chain security, regulatory compliance, and vendor due diligence for public-sector deployments in healthcare and other services. According to The Guardian, the incident underscores the need for code escrow, third-party security audits, and strict incident response SLAs when procuring foundation model services, especially for African government partnerships that may rely on Claude for language processing, content moderation, and decision support. As reported by The Guardian, organizations should reassess data residency, key management, and model governance controls to mitigate IP theft, prompt injection vectors, and downstream compromise in mission-critical use cases.