1. Help Center
  2. Troubleshooting & FAQs
  3. Artificial Intelligence & Machine Learning

What are Edgescan AI Insights?

Edgescan AI Insights are observations on your data as a whole, with regards to specific questions

Version Number: v1.0.1

Published Date: 11 Jun 2024

____________________________________________________________________________

Edgescan AI Insights allows you to ask specific questions about your data. These can be tailored to your use cases.

It should be noted that none of your data will be sent to any third party LLM for processing. We have educated LLM's around the meta-data that is associated with the vulnerabilities and gather insights based on this data. This may be risk rating, public exploits, exposure of asset and your own risk tolerance.

Edgescan AI Insights was released with support for the following questions:

  • Is my Organization vulnerable to ransomware attacks?
    • Which assets or categories of assets in this customer's organization are most susceptible to ransomware attacks - which vulnerability types should they be most worried about?
  • Remediation Priority
    • Outline the remediation prioirities for this organization in a markdown table by vulnerability mentioning the affected assets.
      • Limit the number of rows to the top 10 vulnerabilities based on EXF score
  • Compliance Advice
    • Does this customer have any vulnerabilities that would affect their data compliance certifications? Which assets are affected?
  • Training Advice
    • Does this organization have technical areas where they are struggling? What training could they benefit from?
  • Exploits
    • Does exploit code exist for any of the vulnerabilities on this organization?

As of today, Edgescan is using Amazon Bedrock and a tuned version of Anthropics Claude to infer insights.

The data that gets presented back to you is an AI created insight with parameters tuned and tailored by the Edgescan team for maximum efficiency. Keep in mind that these are AI created, so there is a small chance at making a mistake.