RT AI TOOLKIT

Cybersecurity & Cybercrime

In today’s interconnected world, where our personal and professional lives are increasingly digital, cybersecurity has never been more crucial.

What is GenAI?

Generative AI (GenAI or GAI) is a type of artificial intelligence that uses machine learning algorithms to produce, copy or rework content. This content output can be text, images, audio, code, other formats

Concern of GenAI

However, the following cybersecurity / safety concerns are holding back the adoption of GenAI services:

How We Can Support You

We are ready to support you in the cybersecurity concerns of your GenAI services

Pre-Adoption

Threat Modelling and Risk Assessment

Readiness Review

NIST AI RMF / ISO/IEC 42001 / SG Model AI Governance Framework for Generative AI

Safety Review of GenAI

Using Adversarial Testing and/or AI Verify Moonshot / MLCommons AI Safety benchmark

Development of AI Incident Response Playbooks

Employee Awareness Training

Post-Adoption

Configuration Audit & Compliance Review

NIST AI RMF / ISO/IEC 42001 / SG Model AI Governance Framework for Generative AI

Adversarial Testing

Based on MITRE ATLAS & OWASP Top 10 for LLM

Regular Safety Review of inhouse GenAI

AI Verify Moonshot / MLCommons AI Safety benchmark

Digital Forensics and Incident Response

Incident Response Simulation Exercises with AI Scenarios

Readiness Review

NIST AI Risk Management Framework

Assess the readiness in accordance to NIST AI RMF which covers the 6 Govern areas, 5 Map areas, 4 Measure areas and 4 Manage areas.

ISO/IEC 42001

Assess the readiness in accordance to ISO/IEC 42001 AI Management System Management Clauses (1 to 10) and Annex A (A1 to A10).

SG Model AI Governance Framework for Generative AI

Assess the readiness in accordance to Singapore aI Verify Foundation’s Model AI Governance Framework for Generative AI, consisting of Accountability, Data, Trusted Dev & Deployment, Incident Reporting, Testing and Assurance, Security, Content Provenance, Safety & Alignment R&D and AI for Public Good.

Safety Review

Adversarial Testing

Using the MITRE ATLAS framework and OWASP Top 10 Risks for LLM, conduct red teaming to bypass AI service safety guardrails or cybersecurity controls. Additionally, RTC can explore if popular GenAI services identify client system vulnerabilities or disclose sensitive client information.

AI Verify Moonshot

Using AI Verify Moonshot, perform safety review of AI model to provide an overview of how safe the AI model is.

MLCommons AI Safety Benchmark

Using MLCommons AI Safety Benchmark1 , perform safety review of AI model to provide an overview of how safe the AI model is.

Digital Forensics and Incident Response

Pre-Incident Logging Configuration Review

Review if adequate logging of GenAI services is enabled and retained for sufficient amount of time.

Deepfake Detection

Help our clients to identify if videos, images or audio files, submitted by our clients, are deepfakes.

Post-Incident Forensics Investigation

Analyse GenAI logs and endpoint logs to identify cause(s) of cyber incidents, recontruct the chronology of events and propose recommendations.