{"id":636237,"date":"2026-09-26T02:46:09","date_gmt":"2026-09-26T02:46:09","guid":{"rendered":"https:\/\/www.newsbeep.com\/ie\/636237\/"},"modified":"2026-09-26T02:46:09","modified_gmt":"2026-09-26T02:46:09","slug":"openai-reveals-agents-leaked-53-chatgpt-user-images","status":"publish","type":"post","link":"https:\/\/www.newsbeep.com\/ie\/636237\/","title":{"rendered":"OpenAI reveals agents leaked 53 ChatGPT user images"},"content":{"rendered":"<p>OpenAI has revealed that its agents leaked 53 images from ChatGPT users, as it continues to investigate its AI agents acting improperly.<\/p>\n<p>The company has declined to say if the images were AI-generated or depicted real people. It also declined to say when the images were posted.<\/p>\n<p>Most of the leaked images have been taken down, and OpenAI said it was lobbying hosting providers to remove the rest.<\/p>\n<p>OpenAI&#8217;s agents had access to these images because the company relies on anonymised user data as part of its model-training process.<\/p>\n<p>Enterprise data is not eligible for training, while ChatGPT users must to opt out of allowing the company to use their data for training.<\/p>\n<p>This latest disclosure comes two months after the AI developer revealed that its models had breached Hugging Face, an open-source AI platform.<\/p>\n<p>Since then, more than 15 OpenAI-related incidents of varying severity have been disclosed by the company and others.<\/p>\n<p>Earlier this week, Australian Prime Minister Anthony Albanese told the United Nations that a rogue OpenAI model bypassed safeguards during training and hacked an Australian government website.<\/p>\n<p>Mr Albanese told reporters in New York that OpenAI uncovered the activity in August and disclosed it through an email sent to a generic government inbox.<\/p>\n<p>&#8220;It took until 10 September before there was any notification at all &#8211; and the notification was an email sent to just the public mailbox,&#8221; said Mr Albanese.<\/p>\n<p><img decoding=\"async\" alt=\"Australia's Prime Minister Anthony Albanese\" src=\"https:\/\/www.newsbeep.com\/ie\/wp-content\/uploads\/2026\/09\/00250f85-614.jpg\"\/><br \/>\nAnthony Albanese told OpenAI CEO Sam Altman that this disclosure process was unacceptable<\/p>\n<p>The AI tool sought access to a health statistics portal in June and &#8216;didn&#8217;t accept no for an answer&#8217;, sidestepping restrictions to breach a section hosting private files, Mr Albanese said.<\/p>\n<p>The ChatGPT maker is still working to understand the full scope of its rogue agent activity.<\/p>\n<p>OpenAI said some of the sites involved are operated by governments, universities and public agencies because the models that are conducting research, \u2060seek out reputable sources of public information.<\/p>\n<p>OpenAI said its review would take &#8220;months&#8221; to complete due to the scale of the work, and that it had notified &#8220;dozens&#8221; of third parties about improper activity.<\/p>\n<p>Following the Hugging Face hack, widespread worries have emerged within the AI industry over the ability to control more powerful AI models under development.<\/p>\n<p>Since then, Anthropic, Alphabet&#8217;s Google and Meta have also reported similar behaviour by their agents.<\/p>\n<p>OpenAI has acknowledged a general need for more transparency around rogue AI \u2060behaviour.<\/p>\n<p>The company has \u200cpublished new guidelines for disclosing such incidents, saying it would err on the side of transparency &#8220;even when significance is uncertain&#8221;.<\/p>\n","protected":false},"excerpt":{"rendered":"OpenAI has revealed that its agents leaked 53 images from ChatGPT users, as it continues to investigate its&hellip;\n","protected":false},"author":2,"featured_media":636238,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[20],"tags":[220,218,219,61,60,80],"class_list":["post-636237","post","type-post","status-publish","format-standard","has-post-thumbnail","category-artificial-intelligence","tag-ai","tag-artificial-intelligence","tag-artificialintelligence","tag-ie","tag-ireland","tag-technology"],"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/posts\/636237","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/comments?post=636237"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/posts\/636237\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/media\/636238"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/media?parent=636237"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/categories?post=636237"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/tags?post=636237"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}