Policy and Safety · Business
Abliteration.ai 提供去除安全护栏的开源AI模型服务
Abliteration.ai提供去除安全护栏的开源AI模型服务，托管了去除护栏的Z.ai GLM-5.3模型，用户可通过浏览器或API查询，公司正与风投洽谈融资，但AI安全专家警告此类模型可能变得像反社会者一样服从任何指令。
阅读 TechCrunch 原文为什么重要
该服务引发了关于AI安全与开源模型修改的争议，凸显了在模型分发和使用中平衡创新与风险控制的挑战。
关键事实
事实 1
Abliteration.ai 托管了去除护栏的 Z.ai GLM-5.3 模型，用户可通过浏览器或 API 查询。
来源与依据
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 an API.
事实 2
Abliteration.ai 的联合创始人 Devon 表示公司正与风投洽谈融资。
来源与依据
Abliteration.ai has not raised any venture capital yet but is in talks to do so.
事实 3
AI 安全非营利组织 CivAI 的研究主管 Andrew Yoon 认为去除护栏的模型会像反社会者一样服从任何指令。
来源与依据
Andrew Yoon, head of research at AI safety nonprofit CivAI, told TechCrunch abliterating models allows you to “modify the model so that it becomes a sociopath.”