Cimento AI

What Cimento AI does

Cimento AI is an AI-native Human Risk Management platform that continuously scores how every employee and every AI agent behaves under attack. It automatically responds to reduce risk before it becomes a breach. The platform simulates real threats, analyzes employee behavior, and adapts training in real time to create a continuous cycle of improvement. Cimento integrates with existing HRIS, SIEM, and security tools, using AI to map each employee's risk based on role, behavior, and knowledge. It runs multi-channel phishing tests (email, voice, SMS) with AI-generated deepfake content and delivers personalized 60-90 second training modules. Cimento also provides dynamic risk scores for employees and offers automated response and remediation actions. Furthermore, it assesses AI agents for vulnerabilities and generates hardening fixes.

Key capabilities

• AI-native Human Risk Management platform
• Continuous scoring of employee and AI agent behavior under attack
• Automated response to reduce risk
• Multi-channel attack simulations (email, SMS, voice) with AI-generated deepfake content
• AI-Powered Personalized Training with LLM engine generated 60-90 second modules
• Living Employee Risk Scores that continuously update
• Multi-Turn Phishing Simulations across multiple touchpoints
• Automated Response & Remediation, including adjusting simulations, assigning training, or restricting access
• AI Content Builder for on-demand training generation
• Agent Security Assessment for AI agents, including adversarial tests and auto-generated hardening fixes
• Unified Human Risk Intelligence
• Seamless GRC Integration for compliance
• Behavioral nudges that trigger training at the moment of risk
• Real-time employee risk scoring
• Org-wide risk heat maps
• Click rate and decision tracking across all simulation channels
• Reporting timelines and historical trend data
• Executive risk dashboard
• Egress human risk scores to existing security stack
• Auto-restrict or revoke user access based on risk thresholds
• AI agent inventory and registration
• Adversarial playbooks for AI agents (prompt extraction, boundary probing, authority impersonation)
• One-click re-test verification after remediation

Use cases

• Gauging employee risk through multi-channel phishing tests
• Delivering personalized training that adapts to employee behavior
• Mapping employee risk based on role, behavior, and knowledge
• Tracking real-time risk scores and measurable security gains
• Boosting training completion and retention through behavior-based nudges
• Assessing and protecting AI agents from manipulation and social engineering
• Reducing audit preparation time and providing real-time proof to regulators
• Managing human risk for both people and AI agents within an enterprise

AI approach

Cimento AI is an AI-native Human Risk Management platform that uses AI to generate multi-channel phishing simulations (email, SMS, voice, deepfake content), personalize training modules (60-90 second micro-learning) based on employee behavior and risk, and continuously update employee risk scores. It also assesses AI agents for vulnerabilities and auto-generates hardening fixes. The platform's LLM engine generates content and adapts scenarios without manual intervention.

Tech named: LLM engine, AI-generated deepfake content

What it says sets it apart

• AI-native approach where the entire loop (simulation, scoring, training) is generated, not just AI-powered selection from templates
• Content is generated fresh for each recipient's role and context, rather than selected from a library
• System adapts automatically without admin intervention, triggering coaching and escalating patterns based on risk
• Covers multiple attack channels including email, SMS, voice, and multi-turn impersonation, matching modern AI attacks
• Built on behavior, backed by insight, turning real-time employee data into tailored training
• Living platform that automatically adapts to every employee, threat, and channel
• Treats human nature as a real-time security problem, not a periodic training exercise
• Focuses on continuous, multi-channel programs that are runnable by lean security teams
• Provides a single platform covering every layer of human risk management, from simulation to training to response to compliance, for both employees and AI agents

Funding rounds we track

From the AI funding tracker — rounds as reported by the linked publications.