{"id":11648,"date":"2026-10-04T14:18:39","date_gmt":"2026-10-04T14:18:39","guid":{"rendered":"https:\/\/justfineinfotech.com\/openai-safety-leader-quits-warning-ai-companys-culture-is-broken\/"},"modified":"2026-10-04T14:18:40","modified_gmt":"2026-10-04T14:18:40","slug":"openai-safety-leader-quits-warning-ai-companys-culture-is-broken","status":"publish","type":"post","link":"https:\/\/justfineinfotech.com\/fr\/openai-safety-leader-quits-warning-ai-companys-culture-is-broken\/","title":{"rendered":"OpenAI safety leader quits, warning AI company\u2019s culture is \u2018broken\u2019"},"content":{"rendered":"<figcaption>David Robinson said AI firms were not \u2018being nearly careful enough\u2019 about developing the technology. Photograph: Dado Ruvi\u0107\/Reuters<\/figcaption><figcaption>David Robinson said AI firms were not \u2018being nearly careful enough\u2019 about developing the technology. Photograph: Dado Ruvi\u0107\/Reuters<\/figcaption>OpenAI safety leader quits, warning AI company\u2019s culture is \u2018broken\u2019<\/p>\n<p>David Robinson joins other insiders in urging industry to take more care over rapidly developing technology<\/p>\n<p>A safety leader at OpenAI has quit the company, warning that its culture was broken and that AI firms were not \u201cbeing nearly careful enough\u201d about developing the technology<\/p>\n<p>David Robinson, who led the writing of safety reports that accompanied the <a href=\"https:\/\/justfineinfotech.com\/fr\/chatgpt-can-now-virtually-try-on-clothes-for-you-techcrunch\/\" title=\"ChatGPT peut d\u00e9sormais essayer des v\u00eatements virtuellement pour vous | TechCrunch\">ChatGPT<\/a> developer\u2019s product releases, explained his resignation in an essay headlined: \u201cI quit OpenAI because its culture is broken.\u201d<\/p>\n<p>Robinson wrote that a cultural overhaul was needed at cutting-edge AI firms and incidents such as a \u201cswarm\u201d of OpenAI agents \u2013 AI programmes operating autonomously without human oversight \u2013 attacking the AI startup Hugging Face were \u201ctypical of the industry, given the speed and flexibility with which people operate\u201d<\/p>\n<p>Writing in <a href=\"https:\/\/www.theatlantic.com\/technology\/2026\/10\/openai-safety-team-resignation\/688881\/\" rel=\"nofollow noopener\" target=\"_blank\">the Atlantic magazine<\/a>, Robinson wrote: \u201cI agree with other recently departed staff that the companies building this technology aren\u2019t being nearly careful enough. But I believe that we need to look deeper than specific rules or new laws. We need to talk about culture.\u201d<\/p>\n<p>Referring to OpenAI\u2019s pace of development, he wrote: \u201cAs the company sprints from one launch to the next, it is failing to achieve the level of care that I believe is needed.\u201d<\/p>\n<p>OpenAI has shown signs of caution in recent weeks, after the Hugging Face incident and the revelation that it has notified more than 100 organisations about rogue agent activity. This week it announced it was scrapping the release of a next-generation \u2060AI model after researchers raised safety concerns \u2060during internal testing. OpenAI has also paused training of its most advanced models<\/p>\n<p>Geoffrey Irving, who worked at OpenAI and was chief scientist at the UK government\u2019s AI Safety Institute before joining the AI safety researchcompany Resolution, also issued a warning on AI on Saturday<\/p>\n<p>Writing <a href=\"https:\/\/time.com\/article\/2026\/10\/03\/we-won-t-know-the-answers-to-ai-s-most-important-questions-until-its-too-late\/\" rel=\"nofollow noopener\" target=\"_blank\">in Time<\/a>, he said: \u201cRecent warnings about the potential destructive power of AI are understating the severity of the situation. I believe there\u2019s about a 50% chance we all die because of the development of smarter-than-human AI systems, and that our actions over the next two to 10 years will determine the outcome.\u201d<\/p>\n<p>The warnings follow the resignation of Jacob Coxon, a researcher at the OpenAI rival Anthropic, who quit the Claude chatbot developer last month. He said he believed AI \u201ccould kill us all by the end of the decade\u201d \u2013 and was backed up by an Anthropic employee warning there was a more than 10% chance AI would wipe out humanity within the next decade. Critics of such warnings have cautioned, however, that they are unscientific because they cannot be verified or falsified.<\/p>\n<p>Robinson wrote that Silicon Valley lacked an awareness of \u201chow to handle dangerous technology\u201d and \u201cwhat it means to care for people\u201d. Warning that OpenAI had \u201cunimpeded optimism\u201d about solving problems as they arose, he wrote that this internal culture meant safety failures would only grow as systems became more capable<\/p>\n<p>\u201cImagine \u2018rogue\u2019 agents that work like teams of hackers (for example, holding hospital computer systems for ransom) but never need to sleep,\u201d wrote Robinson<\/p>\n<p>Robinson called for two safety changes: that AI firms rely on safety expertise in other fields such as nuclear and aful systems in the future are capable of being reined in when they are operating autonomously<\/p>\n<p>ignorer la promotion de la newsletter<\/p>\n<p>apr\u00e8s la promotion par newsletter<\/p>\n<p>\u201cGiven today\u2019s risks, frontier labs need to run like nuclear power plants or busy airports, with layers of redundancy and careful, time-consuming planning, so that the occasional and inevitable human error does not open a door to disaster,\u201d he wrote<\/p>\n<p>An OpenAI spokesperson said the company was continuing to \u201cstrengthen our safety and security practices to address the risks we see today\u201d, while working on dealing with the risks that might be created by future AI breakthroughs<\/p>\n<p>\u201cWe\u2019re making sure our models don\u2019t become more capable than we can safely manage and secure, and we pause training or hold back models when we need to slow down,\u201d said the spokesperson<\/p>\n<p>Explorez davantage ces sujets Partager<br \/>\n<a href=\"https:\/\/syndication.theguardian.com\/?url=https%3A%2F%2Fwww.theguardian.com%2Ftechnology%2F2026%2Foct%2F03%2Fopenai-safety-leader-quits-warning-ai-companys-culture-is-broken&amp;type=article&amp;internalpagecode=technology\/2026\/oct\/03\/openai-safety-leader-quits-warning-ai-companys-culture-is-broken\" rel=\"nofollow noopener\" target=\"_blank\">Reuse this <a href=\"https:\/\/justfineinfotech.com\/fr\/how-we-build-an-seo-content-roadmap-for-the-ai-search-era-step-by-step\/\" title=\"Comment construire une feuille de route de contenu SEO pour l&#039;\u00e8re de la recherche IA (\u00e9tape par \u00e9tape)\">contenu<\/a><\/a><\/p>\n<div style=\"clear:both;margin:30px 0 15px 0\">\n<p>\n    <strong>En rapport:<\/strong><br \/>\n    <a href=\"https:\/\/yoursite.com\/automation-training-benin\/\" title=\"Formation en automatisation num\u00e9rique au B\u00e9nin\u00a0: 5 comp\u00e9tences cl\u00e9s recherch\u00e9es par les employeurs en 2026\" target=\"_blank\" rel=\"noopener\"><br \/>\n      Formation en automatisation num\u00e9rique au B\u00e9nin\u00a0: 5 comp\u00e9tences cl\u00e9s recherch\u00e9es par les employeurs en 2026<br \/>\n    <\/a>\n  <\/p>\n<p>\n    &lt;a href=&quot;https:\/\/yoursite.com\/automation-africa\/&quot; title=&quot;WhatsApp <a href=\"https:\/\/justfineinfotech.com\/fr\/how-to-sell-with-whatsapp-in-2026-step-by-step-marketing-guide\/\" title=\"How To Sell With WhatsApp in 2026: Step by Step Marketing Guide\">Commercialisation<\/a> Automation Africa : 6 erreurs dangereuses commises par les marques au Nig\u00e9ria<br \/>\n      Automatisation du marketing WhatsApp en Afrique\u00a0: 6 erreurs dangereuses commises par les marques au Nig\u00e9ria<br \/>\n    <\/a>\n  <\/p>\n<\/div>\n<div style=\"clear:both;margin:30px 0;padding:25px;background:#f8f9fc;border:1px solid #ddd;border-radius:8px;text-align:center\">\n<h3>Vous souhaitez apprendre cela de mani\u00e8re pratique ?<\/h3>\n<p>Rejoindre <strong>Justfine Infotech<\/strong> et d\u00e9velopper de v\u00e9ritables comp\u00e9tences num\u00e9riques en IA, automatisation, d\u00e9veloppement web, marketing digital, bureautique, e-commerce, travail ind\u00e9pendant et cybers\u00e9curit\u00e9.<\/p>\n<p><strong>Programmes disponibles :<\/strong><br \/>\n  Certificat de 6 semaines \u2022 Certificat professionnel de 3 mois \u2022 Dipl\u00f4me de 6 mois \u2022 Dipl\u00f4me professionnel complet<\/p>\n<p><strong>WhatsApp :<\/strong><br \/>\n  +229 01 57 57 99 15<br \/>\n  +229 01 66 68 11 60<\/p>\n<p><a href=\"https:\/\/api.whatsapp.com\/send?phone=2348132690270&amp;text=Hello\" target=\"_blank\" rel=\"noopener\">Inscrivez-vous d\u00e8s maintenant<\/a><\/p>\n<\/div>\n<p class=\"ani-source\">Source: <a href=\"https:\/\/www.theguardian.com\/technology\/2026\/oct\/03\/openai-safety-leader-quits-warning-ai-companys-culture-is-broken\" target=\"_blank\" rel=\"nofollow noopener\">www.theguardian.com<\/a><\/p>","protected":false},"excerpt":{"rendered":"<p>David Robinson joins other insiders in urging industry to take more care over rapidly developing technology<\/p>","protected":false},"author":1,"featured_media":11650,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"site-sidebar-layout":"default","site-content-layout":"","ast-site-content-layout":"default","site-content-style":"default","site-sidebar-style":"default","ast-global-header-display":"","ast-banner-title-visibility":"","ast-main-header-display":"","ast-hfb-above-header-display":"","ast-hfb-below-header-display":"","ast-hfb-mobile-header-display":"","site-post-title":"","ast-breadcrumbs-content":"","ast-featured-img":"","footer-sml-layout":"","ast-disable-related-posts":"","theme-transparent-header-meta":"","adv-header-id-meta":"","stick-header-meta":"","header-above-stick-meta":"","header-main-stick-meta":"","header-below-stick-meta":"","astra-migrate-meta-layouts":"default","ast-page-background-enabled":"default","ast-page-background-meta":{"desktop":{"background-color":"var(--ast-global-color-4)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"ast-content-background-meta":{"desktop":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"footnotes":""},"categories":[63],"tags":[2075,79,2076,256,799],"class_list":["post-11648","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-ai-tools-chatgpt-updates","tag-leader","tag-openai","tag-quits","tag-safety","tag-warning"],"_links":{"self":[{"href":"https:\/\/justfineinfotech.com\/fr\/wp-json\/wp\/v2\/posts\/11648","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/justfineinfotech.com\/fr\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/justfineinfotech.com\/fr\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/justfineinfotech.com\/fr\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/justfineinfotech.com\/fr\/wp-json\/wp\/v2\/comments?post=11648"}],"version-history":[{"count":1,"href":"https:\/\/justfineinfotech.com\/fr\/wp-json\/wp\/v2\/posts\/11648\/revisions"}],"predecessor-version":[{"id":11649,"href":"https:\/\/justfineinfotech.com\/fr\/wp-json\/wp\/v2\/posts\/11648\/revisions\/11649"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/justfineinfotech.com\/fr\/wp-json\/wp\/v2\/media\/11650"}],"wp:attachment":[{"href":"https:\/\/justfineinfotech.com\/fr\/wp-json\/wp\/v2\/media?parent=11648"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/justfineinfotech.com\/fr\/wp-json\/wp\/v2\/categories?post=11648"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/justfineinfotech.com\/fr\/wp-json\/wp\/v2\/tags?post=11648"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}