From boardroom to engineering floor,the AI security trainingyou can complete in 2 days
Fully aligned with OWASP, NIST, ISO/IEC 42001, and Japan's METI AI Business Guidelines. Five simulated incident exercises—not lectures—internalize what really matters. Japan's first tiered curriculum for executives and practitioners.
The invisible threats riding alongside AI adoption
Behind the convenience of ChatGPT, Copilot, and Claude, incidents are shaking enterprises worldwide.
$25.6M
Arup loss
A deepfake CFO video conference deceived staff into making 15 wire transfers totaling $25.6M (January 2024).
71%
Shadow AI usage
71% of knowledge workers use unauthorized AI for work. 68% use ChatGPT through personal accounts (Reco 2025).
€35M
EU AI Act max fine
Up to 7% of global revenue or €35M. High-risk AI obligations enforced from August 2026.
#3
IPA Top 10 Threats 2026
“Cyber risks around AI use” debuts in Japan's annual top 10 information security threats at rank 3.
Samsung
Engineers pasted source code into ChatGPT; the company banned internal ChatGPT use.
Air Canada
A chatbot's incorrect advice created the first legal liability precedent.
EchoLeak (CVE-2025-32711)
Commands hidden in emails caused Copilot to auto-exfiltrate confidential data.
Slopsquatting
20% of AI-hallucinated packages don't exist—poisoning npm/PyPI supply chains.
Executive DAY and Practitioner DAY—two days, split by role
“Strategic framing for leaders, hands-on implementation for engineers.” Separate curricula optimized per role, anchored to the same shared frameworks.
Executive DAY
6 hours
対象 / Audience
Executives, board members, CISO, CDO, CAIO, Legal, Risk leads
Translate AI risks into management language and leave with decision-making frameworks for governance, compliance, and investment.
- Feel the business impact through an AI incident timeline
- Regulatory map including EU AI Act 2026-08
- Executive summary of OWASP Top 10 for LLM
- Role-play incident response exercise
- Build your 1-year AI governance roadmap
Practitioner DAY
7 hours
対象 / Audience
DX teams, IT, SOC, app dev, data science, SecOps
From prompt injection to RAG defense to red teaming—a hands-on day that builds the skills you actually need.
- Hands-on walkthrough of all OWASP LLM Top 10 2025 items
- Direct, indirect, and multi-modal prompt injection exercises
- Reading real-world attacks through MITRE ATLAS
- Implementing Guardrails and RAG defenses
- Red teaming exercises (Promptfoo, DeepTeam)
Full coverage—domestic and international
Most courses focus on just one or two standards. WARP SECURITY spans all four leading frameworks.
OWASP Top 10 for LLM 2025
All 10 generative AI-specific threats: prompt injection, sensitive information disclosure, excessive agency, and more.
NIST AI Risk Management Framework + 600-1
The Govern / Map / Measure / Manage functions and the 12 generative AI risk categories.
ISO/IEC 42001:2023
Master the PDCA basics of an AI Management System (AIMS) with certification readiness in mind.
METI AI Business Guidelines v1.2
The latest Japan domestic guideline by METI/MIC, covering AI agents and physical AI.
We don't stop at theory
Five scenarios modeled on real incidents let participants experience “what if it happened to us.” Executives learn decisions; practitioners learn responses.
Samsung-style information leak
An employee pastes confidential data into ChatGPT. Learn to visualize, control, and replace shadow AI with sanctioned alternatives.
Arup-style deepfake BEC
A deepfake CFO video conference triggers 15 fraudulent wires. Redesign payment flows and layered identity verification.
Air Canada-style hallucination liability
The first legal precedent finding companies responsible for chatbot misstatements. Design disclaimers, RAG, and citation flows.
EchoLeak-style indirect prompt injection
Copilot auto-exfiltrates secrets via crafted email. Hands-on detection of RAG poisoning and output validation.
Slopsquatting-style supply chain
Attackers register packages AI hallucinates. Shut the door on MCP and AI coding entry vectors.
Where we sit in the market
Sitting between executive short briefings and 5-day engineer-only programs. “2 days × specialized × tiered” has been an open gap in Japan—until now.
| WARP SECURITY | Major SI exec training | Overseas specialist (5 days) | Literacy programs | |
|---|---|---|---|---|
| Audience | Both executives & practitioners | Executives | Engineers | All employees |
| Duration | 2 days (splittable) | 2-3 hours | 5 days | Half-day to e-learning |
| Hands-on exercises | 5 simulated scenarios | Discussion-led | Heavy technical | Mostly lectures |
| Framework coverage | OWASP / NIST / ISO 42001 / METI – all four | Selective | OWASP / NIST | METI guidelines |
| Indicative price (per company) | ¥1.0M–1.8M | from ¥500K | from ¥1.3M per person | ¥10K–250K per person |
Plans
Take executives and practitioners together or separately. Combine both for the “Full Pack” with a 10% discount.
Executive DAY
For executives and board members
5-person pack: ¥1,000,000
期間: 1 day (6 hours)
対象: Executives, CISO, Legal, Risk
- AI incident timeline
- Regulatory map (EU AI Act / ISO 42001 / NIST / METI)
- Executive OWASP Top 10
- Incident response role play
- 1-year governance roadmap
- Certificate + digital badge
2-Day Full Pack
Executives + practitioners in sync
5 execs + 20 practitioners (10% off)
期間: 2 days
対象: Full stack: executives to engineers
- Executive DAY + Practitioner DAY
- All 5 scenario exercises
- AI usage policy template
- 6-month Q&A support
- Priority WARP / ZEROCK consultation
- ISO 42001 roadmap quick check
Practitioner DAY
For engineers and operators
20-person pack from ¥1,000,000
期間: 1 day (7 hours)
対象: IT, DX, SecOps, developers
- OWASP LLM Top 10 hands-on
- Prompt injection exercises
- MITRE ATLAS attack reading
- Guardrails / RAG defense build
- Red teaming exercises
- Certificate + digital badge
*Prices exclude tax. Custom pricing is available based on attendee count, delivery format (in-person/online), and customization.
Industry-specific enterprise customizations are available on request.
Frequently Asked Questions
Yes. It's common to separate them by 1-2 weeks and run an internal survey as a between-session assignment.
AI security training, taken seriously
“We know it's risky, we just don't know where to start.” End that state in 2 days. Start with a free consultation tailored to your context.
A 30-minute online consultation. We tailor the curriculum to your AI usage profile.