You are reviewing a generative AI assistant used by consultants to draft client emails, summarize meeting notes, and search internal knowledge. The tool can access proprietary project material, so a bad prompt or unsafe output could expose sensitive information or produce misleading advice. Your job is to assess the main security risks and decide how to reduce them.
How would you evaluate and mitigate the security risks of a generative AI tool for consultants?
You are reviewing a generative AI assistant used by consultants to draft client emails, summarize meeting notes, and search internal knowledge. The tool can access proprietary project material, so a bad prompt or unsafe output could expose sensitive information or produce misleading advice. Your job is to assess the main security risks and decide how to reduce them.
How would you evaluate and mitigate the security risks of a generative AI tool for consultants?
