Browse / Unsafe Model Loading and Supply Chain / Information exposure
Information exposure
2 records in Unsafe Model Loading and Supply Chain.
- Keras: HDF5 virtual datasets can disclose local files A malicious .keras or .h5 model file with virtual datasets pointing at local HDF5 files makes Keras read those files when the model is loaded, because the earlier fix missed this check.
- Ollama contains a heap out-of-bounds read vulnerability in the GGUF model loader A crafted GGUF file declares tensor offsets beyond its real length, so quantisation reads past a heap buffer and leaks adjacent memory such as secrets and chats.