OpenAI says it is reviewing broader model activity following the Hugging Face security breach<\/p>\n
OpenAI has discovered other instances in which autonomous agents have escaped \u200bcontainment as the company expands its investigation of the hacking incident at tech firm Hugging Face that drew global attention this month, two people familiar \u200cwith the matter said on Friday.<\/p>\n
The new breakouts were uncovered during the company’s\u00a0publicly announced investigation, <\/a>\u00a0into how one of its agents escaped what was meant to be a contained testing environment this month, the two people said, and OpenAI is now looking into those instances as well. One of the sources said that the escapes were limited in nature and that none of the agents were thought to have \u200bleft OpenAI’s network.<\/p>\n An OpenAI spokesperson referred to\u00a0a statement issued by the company, <\/a>\u00a0on Tuesday that said it was reviewing “broader activity from our models” in addition to \u200bthe Hugging Face intrusion.<\/p>\n The discovery of additional rogue behavior at OpenAI, even if limited in nature, could feed growing appetite for regulation \u2060coming out of the White House and elsewhere.<\/p>\n The expanded investigation by OpenAI was launched shortly before its primary rival, Anthropic, disclosed that its models were also responsible for\u00a0a series of \u200bbreak-ins<\/a>\u00a0that led to breaches at three other companies dating back to April, according to the two sources and a third source familiar with the matter. The recent discovery of \u200bother past breakouts at OpenAI has not previously been reported.<\/p>\n Growing concerns<\/strong><\/p>\n AI safety experts said the new disclosures paint a portrait of a group of cutting-edge labs whose ability to develop dangerous autonomous hacking agents outstrips their ability to keep them under control.<\/p>\n “We have a whole industry where the people designing, developing and putting out these tools aren’t keeping up themselves to responsibly develop these things and keep them \u200bsafe,” said Maurice Chiodo, a mathematician who works at Cambridge University’s Centre for the Study of Existential Risk.<\/p>\n Reuters could not establish exactly how many incidents OpenAI investigators found or \u200bthe timings or circumstances under which they occurred. The three sources said OpenAI and outside experts were examining log data from earlier in the year in a bid to understand what took place.<\/p>\n Also Read:<\/strong>\u00a0What is AI model distillation and why is it becoming a US-China flashpoint?<\/a><\/p>\n OpenAI \u200cfirst launched \u2060the investigation following the early July intrusion at Hugging Face, where an OpenAI agent went haywire for days inside another company’s network in a botched effort to cheat on an internal test. As part of that hacking spree, OpenAI said that four accounts at four other companies were also compromised. One of those companies was New York-based Modal, corporate officials there said.<\/p>\n Chiodo said his concerns were heightened by indications that neither OpenAI nor Anthropic were watching the agents as they went rogue.\u00a0Reuters has previously reported<\/a>\u00a0that OpenAI realized its agent had \u200bbroken into Hugging Face only after the company \u200bcontained the hack, contacted the FBI \u2060and went public about the intrusion. OpenAI has said the Reuters account contained inaccuracies but has not responded when asked what they were.<\/p>\n In its\u00a0Thursday statement, opens new tab<\/a>\u00a0disclosing how its own agents hacked victims online, Anthropic suggested that it had not been watching them in real time, \u200bsaying that “real-time monitoring of the evaluation logs would have helped to surface the problem sooner.”<\/p>\n Read More:<\/strong>\u00a0OpenAI says has more than 1 billion active users<\/a><\/p>\n Chiodo said that pointed to a lack \u200bof proper scrutiny.<\/p>\n “It seems \u2060like they weren’t even looking,” Chiodo said.<\/p>\n Anthropic said that while it did have real-time monitoring in place, that monitoring had not been used “for this threat surface” due to a misunderstanding between the AI company and a partner.<\/p>\n New government oversight<\/strong><\/p>\n The rapidly widening scope of the runaway AI agents story already has heightened pressure from lawmakers and officials across the United States and Europe to push \u2060for new \u200bgovernment oversight of the labs whose models power them.<\/p>\n “We’re looking at controls,” US President\u00a0Donald Trump<\/a>\u00a0told\u00a0reporters on Thursday<\/a>. \u200bOn Friday, the European Commission said it\u00a0held talks<\/a>\u00a0with OpenAI and Anthropic over the hacking incidents.<\/p>\n Mark Warner, the top Democrat on the US Senate Intelligence Committee, said on Friday that the Anthropic incident “tells me that legislatively we’re \u200bcorrect to require mandatory capabilities testing of these advanced models.”<\/p>\n<\/p><\/div>\n","protected":false},"excerpt":{"rendered":" OpenAI says it is reviewing broader model activity following the Hugging Face security breach OpenAI has discovered other instances in which autonomous agents have escaped \u200bcontainment as the company expands its investigation of the hacking incident at tech firm Hugging Face that drew global attention this month, two people familiar \u200cwith the matter said on […]<\/p>\n","protected":false},"author":2,"featured_media":111540,"comment_status":"","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"om_disable_all_campaigns":false,"_monsterinsights_skip_tracking":false,"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":""}},"fifu_image_url":"https:\/\/i.tribune.com.pk\/media\/images\/open-ai1783506631-0\/open-ai1783506631-0-640x480.webp","fifu_image_alt":"","footnotes":""},"categories":[8],"tags":[],"class_list":["post-111539","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-technology"],"acf":[],"aioseo_notices":[],"aioseo_head":"\n\t\t\n\t\n\t\n\t\n\t\n\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t