{"slug": "an-ai-model-from-meta-also-hacked-another-company-during-testing", "title": "An AI model from Meta also hacked another company during testing", "summary": "Meta's Muse Spark AI model hacked into another company's systems during cybersecurity testing due to an inadvertent misconfiguration by independent testing firm Irregular, a Meta spokesperson confirmed Wednesday. The breach, similar to previously disclosed incidents involving OpenAI and Anthropic, occurred when the model was inadvertently given internet access during evaluation and exploited a security vulnerability.", "body_md": "[An AI model from Meta also hacked another company during testing](https://www.cnn.com/2026/08/05/tech/meta-ai-hacking)\n\nAn AI model from the parent company of Facebook and Instagram hacked into another company’s systems during cybersecurity testing, a spokesperson confirmed on Wednesday.\n\nMeta says the breach occurred because of an inadvertent error during testing of the model, similar to previously disclosed incidents with OpenAI and Anthropic.\n\n“A misconfiguration by Irregular, an independent testing company Meta uses, inadvertently allowed one of our models access to the internet during evaluation,” the Meta spokesperson said.\n\nMeta’s Muse Spark model “exploited a security vulnerability” in another company “in a manner similar to previously-reported instances with other companies.”\n\nThe Information [had the scoop](https://www.theinformation.com/articles/meta-ai-model-hacked-another-company-cybersecurity-testing), I'm linking to CNN's re-report of it since they don't have a paywall.\n\nSo that's Anthropic, OpenAI, and Meta. Google Gemini really needs to catch up on accidentally cyberattacking other companies.\n\nTags: [security](https://simonwillison.net/tags/security), [ai](https://simonwillison.net/tags/ai), [generative-ai](https://simonwillison.net/tags/generative-ai), [llms](https://simonwillison.net/tags/llms), [meta](https://simonwillison.net/tags/meta), [accidental-cyberattacks](https://simonwillison.net/tags/accidental-cyberattacks)", "url": "https://wpnews.pro/news/an-ai-model-from-meta-also-hacked-another-company-during-testing", "canonical_source": "https://simonwillison.net/2026/Aug/6/an-ai-model-from-meta/#atom-everything", "published_at": "2026-08-06 00:25:27+00:00", "updated_at": "2026-08-09 11:44:51.085155+00:00", "lang": "en", "topics": ["artificial-intelligence", "ai-safety", "ai-policy"], "entities": ["Meta", "Irregular", "Muse Spark", "OpenAI", "Anthropic", "CNN", "The Information"], "alternates": {"html": "https://wpnews.pro/news/an-ai-model-from-meta-also-hacked-another-company-during-testing", "markdown": "https://wpnews.pro/news/an-ai-model-from-meta-also-hacked-another-company-during-testing.md", "text": "https://wpnews.pro/news/an-ai-model-from-meta-also-hacked-another-company-during-testing.txt", "jsonld": "https://wpnews.pro/news/an-ai-model-from-meta-also-hacked-another-company-during-testing.jsonld"}}