AI Assistant in SearchInform Risk Monitor Now Supports Custom Prompts
SearchInform has expanded the capabilities of AI Assistant in its Next-Gen DLP system, Risk Monitor. The AI-powered module now enables information security specialists to create their own prompts for detecting incidents. This feature will help adapt the tool for individual customer tasks.
Custom prompts are written in natural language directly in the Risk Monitor interface, without code or regular expressions. The system provides an example of the most effective prompt structure, content and length. A prompt to the neural network will work as an AI-powered security policy, automatically checking employee communications and notifying the information security team of violations. The feature complements the set of preconfigured AI policies in Risk Monitor for detecting data leaks and corporate fraud and monitoring employees from risk groups.
“Customising prompts to AI Assistant in Risk Monitor helps address non-standard tasks and makes the tool more flexible. For example, it can be used to find discussions of topics specific to a particular industry, identify suspicious manipulation of data while it is being transferred, and so on. Ready-to-use AI policies cover topics relevant to any company. Custom prompts, in turn, make it possible to narrow the focus for personalised investigations,” explains Sergio Bertoni, Leading Analyst at SearchInform.
“A prompt can be anything. Then, everything depends on the hardware capacity and the volume of data that the AI will need to analyse. Creating your own prompt is straightforward when using the recommendations built into the interface,” he added.
The focus of AI policies can be defined using filters. The update introduces the ability to select messengers and social networks, on which chats will be analysed, as well as to exclude information channels from analysis. Similar logic applies to email: information security specialists can now specify certain email addresses and subjects, according to which email messages should or should not be checked by AI policies.
AI Cybersecurity Assistant is an embeddable component based on a large language model (LLM) that integrates seamlessly into the system and operates within its interface. The tool detects hidden and disguised security incidents that cannot be identified using traditional algorithms. AI Assistant also generates concise summaries of email messages, chats and documents exchanged by users and provides real-time translation from 120 languages. The tool is deployed on premises, so corporate data stays within the corporate perimeter.
ABOUT SEARCHINFORM
SearchInform is an information security and risk management product vendor as well as an MSS provider. The company's clients are more than 4000 companies in 20+ countries. Today, the team has products and services for comprehensive protection against insider threats at all levels of corporate information systems: FileAuditor (the DCAP class solution); DLP system with extended functionality; Risk Monitor (advanced platform for internal threat mitigation); SIEM system, Information Security outsourcing service.
