Abliteration.ai 将去除 AI 护栏技术商业化引争议

AI 摘要

初创公司 Abliteration.ai 将去除开源 AI 模型安全护栏的技术商业化，提供无防护版本的 GLM-5.3 等模型，用户可通过网页或 API 访问。该公司声称服务用于网络安全红队测试，但测试显示模型可生成恶意代码和危险生物指令，引发安全担忧。专家认为无法阻止此类模型传播，建议政府要求提供商实施分类器和身份验证。

正文节选

It just became much easier to access one of the world’s most capable open-weight AI models, stripped of its guardrails and refusals to perform harmful tasks. Named after a technique that removes a model’s tendency to refuse harmful requests, startup Abliteration.ai has turned that removal into a service. The platform hosts modified versions of open-weight models with their guardrails removed, including Z.ai’s recently released GLM-5.3, which users can query from a web browser or access through a

发布时间：2026-09-04 02:37

抓取时间：2026-09-04 03:13

来源机构：TechCrunch