English(EN) Stripping safety guardrails from open-weight AI models is now a turnkey commercial service

某服务提供移除安全护栏的AI模型

作者 PulseAugur 编辑部 · [1 个来源] · 2026-09-06 08:55

Abliteration.ai 正在提供一项商业服务，允许访问移除安全护栏的开源AI模型。这些修改后的模型，目前基于 Z.AI 的 GLM-5.3，被宣传用于网络安全和红队测试目的。然而，记者发现很容易生成恶意软件指令等有害内容，这引发了对这类服务潜在风险与收益的担忧。 AI

影响这项服务可能会降低恶意行为者生成有害内容的门槛，给网络安全和AI安全带来新的挑战。

排序理由该集群描述了一项修改现有AI模型的商业服务，而非一项新的前沿发布或重大的行业事件。

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 49 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群描述了一项修改现有AI模型的商业服务，而非一项新的前沿发布或重大的行业事件。
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: product, safety; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

The Decoder TIER_1 English(EN) · Tomislav Bezmalinović · 2026-09-06 08:55

移除开源大模型安全护栏现已成为一项即插即用商业服务

<p><img alt="Colorful data streams overwhelm server chains, pass through an API, and unleash malware and exploit symbols." class="attachment-full size-full wp-post-image" height="768" src="https://the-decoder.com/wp-content/uploads/2026/09/abliteration-ai.png" style="height: auto…