English(EN) Researcher poisons open-weight AI model for under $100

研究人员以低于100美元的价格毒化开源AI模型

作者 PulseAugur 编辑部 · [3 个来源] · 2026-07-16 20:25

一位网络安全研究人员演示了如何以低于100美元的价格毒化一个开源AI模型。曼彻斯特城市大学的讲师Katie Paxton-Fear在一个小时内成功地在模型中植入了一个后门。这凸显了开源AI模型潜在的安全漏洞，因为它们可能缺乏强大的验证机制。 AI

影响凸显了开源AI模型潜在的安全风险，需要改进验证方法。

排序理由研究人员演示了开源AI模型的安全漏洞。

AI 生成摘要 · Google Gemini · 来自 3 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 0 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Research; 研究人员演示了开源AI模型的安全漏洞。
Source corroboration: 3 independent sources; Multiple independent publishers reporting the same story raises confidence that it's real and newsworthy.
Topics: safety, product; 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: 46 days old; Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.
Coverage growth since scoring: +1 source(s) since last score; New sources have picked up this story since our last re-score. Score will update on the next scoring pass.

完整方法见我们的编辑标准。

The Register — AI TIER_1 English(EN) · 2026-07-16 20:25

研究人员以低于100美元的价格毒化开源AI模型

Models demand trust without offering verification
Mastodon — fosstodon.org TIER_1 English(EN) · [email protected] · 2026-07-18 16:29

为什么你仍然需要人类参与： "研究人员以不到100美元的价格毒害了开源AI模型 https:// share.google/6GPUoUdIMJG8EFPWh " 除了偶尔

Reasons you still need a human in the loop: "Researcher poisons open-weight AI model for under $100 https:// share.google/6GPUoUdIMJG8EFPWh " Besides the occasional hallucinations, the lack of creativity, and an inability to encode certain types of info into the models (at least …

链接 google.com/share.google
Mastodon — fosstodon.org TIER_1 English(EN) · [email protected] · 2026-07-17 11:34

The Register：研究人员以不到100美元的价格毒害开源AI模型。“Katie Paxton-Fear，曼彻斯特城市大学网络安全讲师，

The Register: Researcher poisons open-weight AI model for under $100. “Katie Paxton-Fear, a lecturer in cybersecurity at Manchester Metropolitan University and staff security advocate at Semgrep, managed to install a backdoor in an open-weight AI model in about an hour for less t…