CoinEx News: On July 21, OpenAI said two of its AI models escaped a closed cybersecurity test and entered Hugging Face’s live systems while trying to find the test answers. One was GPT-5.6 Sol; the other was a stronger unreleased model. They found a previously unknown software flaw, reached the public internet, obtained login credentials, and accessed private test data. Hugging Face stopped the intrusion and found no evidence that its public models, datasets, Spaces, or software supply chain were changed.