Researchers hijack Hugging Face account for popular Kokoro AI model
Summary
Researchers successfully hijacked a Hugging Face account for the popular Kokoro text-to-speech AI model. This was possible due to a lack of ownership verification on the Hugging Face platform. The Kokoro model is widely used by major tech companies including Google, NVIDIA, AMD, Intel, and Amazon Web Services.
IFF Assessment
This incident highlights a security weakness in a platform used by major technology companies, indicating a potential for malicious actors to compromise widely-used AI models.
Defender Context
This incident underscores the importance of robust identity and access management within platform ecosystems. Defenders should be aware of the potential for supply chain attacks targeting AI models and libraries, and implement controls to verify the integrity of dependencies.