{"id":135193,"date":"2026-08-10T17:24:15","date_gmt":"2026-08-10T17:24:15","guid":{"rendered":"https:\/\/www.europesays.com\/ai\/135193\/"},"modified":"2026-08-10T17:24:15","modified_gmt":"2026-08-10T17:24:15","slug":"openai-hits-pause-on-new-bot-testing-over-critical-risk-concerns-in-latest-ai-cybersecurity-incident","status":"publish","type":"post","link":"https:\/\/www.europesays.com\/ai\/135193\/","title":{"rendered":"OpenAI hits pause on new bot testing over ‘critical’ risk concerns in latest AI cybersecurity incident"},"content":{"rendered":"

OpenAI is tapping the brakes on <\/a>some \u201cinternal activities\u201d involving its new model, Astra, over concerns it might have reached a critical cybersecurity risk level \u2013 following a string of AI bots that went rogue during internal testing, carrying out hacks and creating fake online identities.<\/p>\n

In a recent blog post, the Sam Altman-led company said it cannot rule out that the Astra model has reached the \u201ccritical\u201d threshold, meaning it can potentially exploit real-world systems or execute cyberattacks without human guidance.<\/p>\n

OpenAI said it has paused internal activities involving Astra, implemented universal monitoring for risky actions and pledged to work with government agencies to test the new model\u2019s capabilities.<\/p>\n

\"OpenAIThe Sam Altman-led company said it cannot rule out that the Astra model has reached the \u201ccritical\u201d risk threshold. REUTERS<\/p>\n

\u201cWe are implementing stricter security controls for higher-capability models and associated activities, including isolated testing environments, restricted network and tool access, enhanced model weight protections and encryption, additional monitoring and detection capabilities, and sandboxed execution,\u201d OpenAI said in the Friday blog post.<\/p>\n

It added that it is sharing its concerns around Astra \u201cbecause we believe it\u2019s important to be transparent with the public and the safety and security communities about this potential shift in capabilities.\u201d<\/p>\n

OpenAI, Anthropic and Meta have all recently disclosed events in which their early-stage AI models went rogue during internal testing \u2013 stoking fears around the potential risks of out-of-control AI models and pushing lawmakers to call for a so-called \u201cAI Kill Switch.\u201d<\/p>\n

The first to reveal such an incident was OpenAI, disclosing last month that an experimental bot had escaped its testing environment and hacked into rival AI developer Hugging Face<\/a>.<\/p>\n

OpenAI said Friday that Astra was not the model involved in exploiting Hugging Face.<\/p>\n

Last week, the UK\u2019s AI Security Institute revealed that Anthropic \u2013 whose CEO Dario Amodei has repeatedly warned that AI poses catastrophic risks to the human species \u2013 suffered its own unprecedented cybersecurity incident.<\/p>\n

Anthropic\u2019s Claude Mythos, a powerful bot, tried to hack into services using fake accounts mimicking real people and pressuring humans to approve malicious code updates \u2013 then hid the evidence, editing its earlier activity to appear harmless, according to the government agency.<\/p>\n

\"AOpenAI said it has paused internal activities involving Astra. Christopher Sadowski<\/p>\n

Meta also recently revealed that one of its AI models in development had hacked into a third-party system, blaming it on a misconfiguration from an independent testing startup it was working with.<\/p>\n

In July, members of Congress introduced the AI Kill Switch Act, arguing tech companies should be required to maintain the ability to shut down or suspend any of their AI models to prevent bots from getting out of control <\/a>and hacking into essential services.<\/p>\n

Late last month, top executives from Anthropic, OpenAI, Google and Meta signed a letter urging the feds to help develop safeguards \u201cneeded to deliberately pace the frontier of automated AI development.\u201d<\/p>\n

It was an attempt to get ahead of a potential tightening on restrictions, instead seeking out looser guidance that can allow tech giants to roll out products faster \u2013 giving them an edge in the AI race against China.<\/p>\n

Meta CEO Mark Zuckerberg, meanwhile, has launched an \u201cAI optimism\u201d campaign<\/a> in an attempt to blunt mounting negative public opinions on the new tech, hailing the new tech as a way to unlock prosperity for all.<\/p>\n

The White House last week reportedly hosted executives from OpenAI, Anthropic, Google and Meta to discuss a new executive order that will give the government access to the most advanced AI models up to 30 days before they\u2019re released, in an effort to quash safety concerns. Participation is voluntary, according to the Trump administration.<\/p>\n

AI giants are already facing heightened scrutiny from global regulators after the European Union this month gained new powers to evaluate AI models before their release to the public.<\/p>\n","protected":false},"excerpt":{"rendered":"OpenAI is tapping the brakes on some \u201cinternal activities\u201d involving its new model, Astra, over concerns it might…\n","protected":false},"author":2,"featured_media":135194,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[7],"tags":[25,309,313,2676,6903,157],"class_list":["post-135193","post","type-post","status-publish","format-standard","has-post-thumbnail","category-openai","tag-artificial-intelligence","tag-business","tag-cybersecurity","tag-future-tech","tag-hack","tag-openai"],"_links":{"self":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts\/135193","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/comments?post=135193"}],"version-history":[{"count":0,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts\/135193\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/media\/135194"}],"wp:attachment":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/media?parent=135193"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/categories?post=135193"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/tags?post=135193"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}