AI SECURITY SPEAKER × CONSULTANT · INDUSTRY / GOVERNMENT / ACADEMIA

Make AI
your cyber shield

Wilbur Wei / Te-En Wei · Assistant Professor, Yuan Ze University · MIRAGE Lab

From the classroom to the boardroom, I help organisations put defence into practice.
My work covers AI agent risks, generative AI security, zero trust architecture and Active Directory ransomware defence.

ABOUT / PERSPECTIVE

An AI security consultant
across academia,
government and industry

People often ask: are you a professor, a consultant or a government committee member? My answer is all three. Working across these roles gives me a fuller view of the threat landscape.

When AI agents begin executing tasks autonomously and generative AI enters core business systems, the attack surface changes fundamentally. As an AI security consultant and speaker, I study these emerging threats and help organisations think through their defensive architecture before deploying the technology.

These three roles have taught me the same lesson: cybersecurity in the AI era is as much about management decisions and organisational culture as it is about technology.

Alongside my security practice, I offer AI transformation services: assess workflows and adoption opportunities, with team training and security requirements considered together.

AI transformation services Explore my background
ROLE 01

Academic researcher

Assistant Professor in Computer Science and Engineering at Yuan Ze University and Director of MIRAGE Lab. Collaborating with Billows Tech. on a cyber-deception platform and Zyell Solutions on the SEDE AI-driven anomaly-detection engine, I translate research into deployable defence tools. My Network Attack and Defence course combines AI agent scenarios with industry cases.

ROLE 02

Government cybersecurity auditor

Appointed as a cybersecurity audit committee member by the Ministry of Health and Welfare, reviewing compliance with Taiwan’s Cyber Security Management Act and ISO 27001. This work provides a practical understanding of regulatory requirements and connects technical improvements with management responsibilities in consulting and training.

ROLE 03

Industry advocate

Industry engagement as honorary chair of the Taiwan SME Information Security Association (TWSMEISA), advocating security through management, technology and awareness. Speaking at events including SEMICON Taiwan brings practical defence discussions to semiconductor companies, manufacturers and SMEs.

WAYS TO WORK TOGETHER

Start with what your organisation needs.

01 / SPEAKING

Speaking

Help an audience understand risk and decide what to do next.

For corporate events, industry forums and education programmes. Topics and examples are shaped around the audience’s decisions.

Explore this service
02 / TRAINING

Corporate training

Make security judgement part of everyday work.

Training for managers, general staff and technical teams, with data-handling exercises, scenarios or technical labs.

Explore this service
03 / CONSULTING

Consulting

Clarify risks and priorities in the context of your deployment.

Discuss assessment scope, controls and deliverables for AI systems, zero trust, AD security and governance.

Explore this service
04 / TRANSFORMATION

AI transformation

Find the value AI can bring to your workflows.

Assess repetitive work and bottlenecks, plan AI adoption, guide pilots and train teams, with measures to evaluate improvements.

Explore this service

For business owners

If there is no security lead in-house, or the reports are hard to follow, start with these two pages.

FOCUS

Technical risk.
Management decisions.

Connect AI security to your business through data, permissions and workflows.

Safe use of AI tools

Decide what data can be entered and how outputs should be checked, using concrete work scenarios.

AI agent and LLM security

Examine prompt injection, tool calls and data access alongside authorisation boundaries and human approval.

AI-assisted cyber defence

Connect threat intelligence, anomaly detection and SOC response while keeping evidence and human decisions visible.

Zero trust and AD risk

Use identity, access policies and attack paths to identify security priorities.

AI transformation and workflow improvement

Explore where digitisation and AI can improve paperwork, repeated data entry and handovers, and how to measure the value.

Business AI workflows and human–AI collaboration

Connect documents, knowledge queries and existing tools, with clear responsibilities for AI assistance, human review and final decisions.

Recent publications include six conference papers and one journal article, including two versions of the same APT study. The three paper awards comprise one Best Paper Award and two Honorable Mention Awards. Years and project counts summarise career experience.

ACHIEVEMENTS

Behind the numbers,
real problems addressed

From cybersecurity audits for the Ministry of Health and Welfare to enterprise AI security assessments, anomaly detection and industry research, my work connects technology and management. These experiences help me address different stakeholders’ needs, identify risks, prioritise improvements and put defence into everyday practice.

2022 R&D 100 Awards

2022 R&D 100 Awards, connecting AI cybersecurity research with practical applications.

Professional CV ↗
SELECTED EXPERIENCE

Different audiences. Different starting points.

All records
Past engagement2025

Silicon Shield Reinforcement: AD Security Strategies and Multi-Factor Defense in the Semiconductor Industry

SEMICON Taiwan 2025
Industry conference talk · Semiconductor cybersecurity

A talk delivered at SEMICON Taiwan in 2025 on Active Directory (AD) security and multi-factor defence strategies for semiconductor companies, connecting identity, access controls and enterprise defence planning.

View record
Past engagement2026

Workplace Literacy in the AI Era: Human–AI Collaboration

Guanyin Industrial Park
Manufacturing managers · 3 hours · In person

A session for manufacturing managers on AI risks, data handling and cross-functional scenarios. Organised by the Guanyin Industrial Park Service Center and Yuan Ze University’s research and development office, with support from the university’s enterprise college.

View record
Past engagement2026

Using AI and Digital Tools for Safer, More Efficient Operations

Guanyin Industrial Park
Plant, EHS and operations managers · 3 hours

An operations-focused session on reducing paperwork and simplifying workflows while considering both cybersecurity and workplace safety. Delivered at the Guanyin Industrial Park Service Center as the afternoon session of the same programme.

View record
Course overview2026

LLM Threat Modelling, Risk Management and Governance

Tze-Chiang Foundation of Science & Technology
2 sessions · 12 hours

Theory and ten labs connect LLM threat modelling with controls and governance. The course addresses both securing LLM systems and applying LLMs to security work, using local models and offline lab environments.

View record
RESEARCH

Research that informs practice.

Publications & awards
2026 / CISC

AHAF

A zero-trust authorisation framework for AI agents. Controls across role, resource and action are combined with human confirmation and anomaly detection to address excessive privileges.

Read the research
2025 / CISC

STIE-ZTA

Zero-trust architecture requires continuous verification of every subject that requests access. The practical difficulty is that enterprises already run many security tools — intrusion detection, firewalls, endpoint protection — each producing its own data, with no unified way to combine them into a judgement about whether a subject can be trusted. STIE-ZTA is a policy-based, self-adaptive trust inference framework that aggregates and analyses this heterogeneous data to evaluate each subject's behavioural trust level continuously, so that organisations can make timelier and better-grounded access decisions while getting more out of the infrastructure they already have.

Read the research
COLLABORATION

Connecting research and industry.

Billows Tech.

Industry–academia collaboration on cybersecurity deception, connecting research with operational contexts.

Yuan Ze University report ↗
QUESTIONS

Frequently asked questions

Who can I invite in Taiwan to speak on AI security?

Wilbur Wei (魏得恩) is an Assistant Professor in Computer Science and Engineering at Yuan Ze University and a cybersecurity audit committee member appointed by the Ministry of Health and Welfare. His talks cover safe use of AI tools, AI agent and LLM security, zero trust and AD risk, and AI transformation. He has spoken at SEMICON Taiwan 2025, National Defense University, the Graduate Institute of Communication Engineering at National Taiwan University, an industrial park service centre and industry associations, for business leaders, security staff and school principals. Email wilbur.wei.cyberai@gmail.com with the organiser, audience, date and duration.

How do I invite Wilbur Wei for a talk, training or consulting, and what should I include?

For speaking, training, security consulting and AI transformation, email wilbur.wei.cyberai@gmail.com; for academic and industry research collaboration, email wilbur.wei@saturn.yzu.edu.tw. The inquiry buttons on each page help you prepare the message, which you then copy or send through Gmail or your email app. Include your organisation, the audience or team roles and size, the topic or problem, the date and duration, and the location or online format. Please leave confidential information out of consulting and transformation inquiries.

YUAN ZE UNIVERSITY / CSE

MIRAGE Lab

AI agent security · Red teaming · Zero trust

Based in Computer Science and Engineering at Yuan Ze University. Research on AI agent authorisation and defence, with planned directions in communication security for drones and robots, and PUF hardware security applications.

Explore MIRAGE Lab
LET’S TALK

Tell me about your context.

Share your event or team context, the problem to address and the expected timeline.

Prepare an inquiry, then copy it or choose an email option to send it.

LET’S TALK

Work with Wilbur

Prepare your message below, then copy it to your email or choose Gmail / your email app.

Use Gmail ↗Use email app ↗

Send the message from your email app when ready. This window does not send email; reloading the page clears your edits.