In MONAI 1.6.0, PersistentDataset (monai/data/dataset.py) explicitly rejects the combination track meta=True with weights only=True, forcing users who cache MetaTensors (the default tensor type in MONAI >= 1.0) to run torch.load(hashfile, weights only=False). Related cache helpers in monai/data/utils.py also call pickle.loads on cached content and derive cache keys with hashlib.md5. As a result, a local user with write access to a shared or world-writable cache dir (e.g. /tmp/monai cache, HPC scratch, ~/.cache/monai) can place a malicious pickle file that is deserialized the next time another user's MONAI pipeline reads the cache, resulting in arbitrary code execution in that user's context. All released versions of the monai pip package are affected; no patched version is available as of the advisory.