Google Deploys Gemini AI on Dark Web to Improve Threat Intelligence Accuracy

Google Cloud is continuing to apply generative AI to threat intelligence, aiming to identify signs of attacks more quickly across dark-web forums, malware and leaked data. Dark-web information is vast and often unreliable, making manual analysis time-consuming. Gemini is designed to help cybersecurity teams filter content, cross-check sources and shorten the time between detecting a threat and taking action.

At the RSA Conference in April 2025, Google Cloud announced that it was deploying Gemini AI deep into the dark web to gather intelligence. The system automatically analyzes more than 1 million posts a day and filters signals from tens of millions of pieces of content, achieving 98% accuracy in threat reports. The company also unveiled security-agent automation services and Model Armor to guard against risks in AI applications, including prompt injection and sensitive-data leaks, but did not disclose the amount invested.