ENISA used an OpenAI model to find four flaws in EU code, Politico reports

The EU's cybersecurity agency ENISA used an advanced OpenAI model to identify four security flaws in EU project code, including a high-risk vulnerability that could allow account hijacking. These flaws have been fixed following the discovery. ENISA and CERT-EU conducted the analysis after gaining access to US AI models, including Mythos and GPT-6 Astra, through partnerships with OpenAI.