New AI tool to process classified informations

American firm Anthropic has developed a new custom-built large language model called Claude Gov, designed to handle and process classified data. This AI is akin to other consumer-facing AI models, such as OpenAI’s ChatGPT or Google’s Gemini. However, while these are trained to reject sensitive or confidential information, Claude Gov can process this type of input and can support numerous tasks such as strategic planning, threat detection, and operational support, enhancing intelligence analysis across defence, language, and cybersecurity. Claude Gov is already in use by some undisclosed US national security agencies, which reflects the increasing tendency of government and defence agencies to employ AI to speed up data analysis and improve decision-making.