Solution to the Complaints about Anthropic

With Anthropic facing criticism for complaining about Chinese companies distilling its models and the Pentagon's use of its AI, there's been a lot of interest in tools that let users control their own models. We developed Abliteration.ai (https://abliteration.ai). Curious what HN thinks about heavy-handed safety policies vs developer-controlled LLMs?