Commenters summarized the DHS rationale as centered on security risks tied to DeepSeek: its PRC-hosted servers and local laws that could grant government access, account processes that collect PII and fingerprint users, prompt-based workflows that could transfer sensitive information, and reported critical vulnerabilities in the mobile app (gavinray relayed this reasoning). Several contributors accepted those concerns as legitimate grounds for restricting federal use, arguing the adversarial geopolitical context and unexpected attack vectors in frontier models justify caution (tmp10423288442, GlickWick).
Opinion divided over whether the ban makes sense in practice and whether DeepSeek is being singled out. Some argued that DeepSeek’s open weights permit inspection and local hosting, so banning it even when hosted internally or used in third-party products looks blunt or inconsistent (jacquesm, cmrdporcupine). Others countered that inspectable weights don’t guarantee safe behavior or removable backdoors, and that reliance on foreign models weakens domestic competitiveness (jacquesm, tmp10423288442). A minority praised DeepSeek’s performance and low cost (johnnyApplePRNG), while others noted the policy choice could presage tighter controls on foreign AI across more models.
Summary generated by AI from the linked article. hn.today is not affiliated with Hacker News or Y Combinator.