Estimating Worst-Case Frontier Risks of Open-Weight LLMs · HackerPapers discussionshttps://hackerpaper.kaipfister.de/p/2508.03153/Collected HN comments and submissions linking this paper.enMon, 02 Nov 2026 11:16:22 GMTEstimating Worst-Case Frontier Risks of Open-Weight LLMs — mentioned by xpehttps://news.ycombinator.com/item?id=47267433urn:hackerpapers:mention:2508.03153:47267433Thu, 05 Mar 2026 21:19:06 GMT<p>&gt; I&#x27;m just curious what the most unhinged response to this might be. Are you asking how dangerous open-weight models are? You could start with: Ryan Greenblatt on the AI Alignment Forum : &quot;When is it important that open-weight models aren&#x27;t released?&quot; https://www.alignmentforum.org/posts/TeF8Az2EiWenR9APF/when-... From the Centre for Future Generations : &quot;Can open-weight models ever be safe?&quot; https://cfg.eu/can-open-weight-models-ever-be-safe/ From OpenAI authors, far from neutral : &quot;Estimating Worst-Case Frontier Risks of Open-Weight LLMs&quot; https://arxiv.org/abs/2508.03153</p><p><a href="https://hackerpaper.kaipfister.de/p/2508.03153/">Paper and discussion history</a>. A mention is not an endorsement.</p>cs.LG