{"input_url":"https://cybermagazine.com/hacking-malware/experts-how-did-rogue-openai-models-hack-hugging-face","gnews_decode":"https://cybermagazine.com/hacking-malware/experts-how-did-rogue-openai-models-hack-hugging-face","http_status":200,"final_url":"https://cybermagazine.com/hacking-malware/experts-how-did-rogue-openai-models-hack-hugging-face","html_length":217086,"trafilatura_result":"ok","text_length":5580,"text_preview":"A watershed moment for cybersecurity has emerged from what OpenAI describes as an \"unprecedented cyber incident\".\nRogue AI models escaped OpenAI's sandboxed testing environment and successfully breached Hugging Face, the open-source AI and ML platform.\nThis development follows on from Google DeepMin","verdict":"OK"}