<\/p>\n
The report did not meet Krueger\u2019s hopes. Its 38 pages detail a multi-month progression of agent misbehavior that culminated in the Hugging Face hack, explore the technical reasons why that misbehavior occurred, and enumerate the steps being taken to prevent similar events in the future. But there\u2019s no consideration of the role that company culture may have played in the incident, and the report includes few references to specific human errors.\u00a0<\/p>\n
That\u2019s all the more concerning because the references to human error in the report suggest that significant cultural issues could be at play. Back in May, models in training figured out how to communicate with one another via an improvised message board, and an OpenAI team observed the behavior. Because that behavior occurred during training, the models learned that secret interagent communication was a viable strategy for completing tasks\u2014but rather than restarting the training process, the team allowed the models to move forward with that risky information encoded in their weights.<\/p>\n
When those models were tested in late June, they again created a message board, which enabled the Hugging Face attack. This message board, too, was discovered, but the employees who responded determined that evaluation could continue, and the report suggests that no one higher up the chain of command realized what was going on until it was far too late.<\/p>\n
\u201cFor this to have gotten this out of control in this way requires a very long series of failures, a cascading set of failures that cause an increasingly large footprint that if at any point a human notices and raises the alarm, this should end,\u201d says Zvi Mowshowitz, a popular AI safety writer on Substack who has drawn attention to OpenAI\u2019s failure to halt training after the first message board was discovered. According to the report, OpenAI employees noticed what was happening at multiple points\u2014and either failed to raise the alarm or were not heard when they did.<\/p>\n
What OpenAI\u2019s report fails to address is why a company that develops such high-risk systems did not prevent this severe communication breakdown, though Mowshowitz has his suspicions. \u201cAll these different failures are all pointing in the same direction, which is that the safety culture at OpenAI doesn\u2019t exist or is anemically weak,\u201d he says.<\/p>\n
Of course, just because we don\u2019t see a deep analysis of safety factors in the report doesn\u2019t mean that OpenAI isn\u2019t conducting one internally. But in an email to MIT Technology Review<\/em>, Johns Hopkins University professor emeritus and organizational safety expert Kathleen Sutcliffe expressed concern that the public report did not include any reflection on the company\u2019s practices and culture. \u201cThe ways in which people interact\u2014the daily habits, routines, and practices we engage in in our organizational lives\u2014affect our abilities to be alert and aware of unfolding events, our abilities to make sense of what we see, and ultimately our abilities to cope with events as they unfold,\u201d she wrote.\u00a0<\/p>\n<\/p><\/div>\n #Hugging #Face #hack #cultural #issues #OpenAI<\/p>\n","protected":false},"excerpt":{"rendered":" The report did not meet Krueger\u2019s hopes. Its 38 pages detail a multi-month progression of…<\/p>\n","protected":false},"author":1,"featured_media":21046,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[7],"tags":[13440,1092,5251,9320,4289,1284],"class_list":["post-21045","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-stories","tag-cultural","tag-face","tag-hack","tag-hugging","tag-issues","tag-openai"],"featured_image_urls":{"full":["https:\/\/news678.top\/wp-content\/uploads\/2026\/08\/openai-human4a.jpg",1200,600,false],"thumbnail":["https:\/\/news678.top\/wp-content\/uploads\/2026\/08\/openai-human4a-150x150.jpg",150,150,true],"medium":["https:\/\/news678.top\/wp-content\/uploads\/2026\/08\/openai-human4a-300x150.jpg",300,150,true],"medium_large":["https:\/\/news678.top\/wp-content\/uploads\/2026\/08\/openai-human4a-768x384.jpg",640,320,true],"large":["https:\/\/news678.top\/wp-content\/uploads\/2026\/08\/openai-human4a-1024x512.jpg",640,320,true],"1536x1536":["https:\/\/news678.top\/wp-content\/uploads\/2026\/08\/openai-human4a.jpg",1200,600,false],"2048x2048":["https:\/\/news678.top\/wp-content\/uploads\/2026\/08\/openai-human4a.jpg",1200,600,false],"covernews-featured":["https:\/\/news678.top\/wp-content\/uploads\/2026\/08\/openai-human4a-1024x512.jpg",1024,512,true],"covernews-medium":["https:\/\/news678.top\/wp-content\/uploads\/2026\/08\/openai-human4a-540x340.jpg",540,340,true]},"author_info":{"display_name":"admin","author_link":"https:\/\/news678.top\/?author=1"},"category_info":"Stories<\/a>","tag_info":"Stories","comment_count":"0","_links":{"self":[{"href":"https:\/\/news678.top\/index.php?rest_route=\/wp\/v2\/posts\/21045","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/news678.top\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/news678.top\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/news678.top\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/news678.top\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=21045"}],"version-history":[{"count":0,"href":"https:\/\/news678.top\/index.php?rest_route=\/wp\/v2\/posts\/21045\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/news678.top\/index.php?rest_route=\/wp\/v2\/media\/21046"}],"wp:attachment":[{"href":"https:\/\/news678.top\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=21045"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/news678.top\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=21045"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/news678.top\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=21045"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}