AI SECURITY FOR BUSINESS
I will help your company secure your AI and Large Language Model (LLM) deployments by hands-on workshops, awareness training, professional security assessments and red teaming of your LLM applications, MCP servers and Agentic AI systems.
Schedule a consultation →
SERVICES
Training for development and security teams. Your engineers will learn how to identify threats, exploit vulnerabilities, and defend against attacks specific to AI systems through hands-on exercises. In addition to classic threats such as prompt injection and jailbreaking, they will also learn how to attack MCP servers and AI agents, as well as how to securely integrate LLMs with your web applications.
AI security awareness training for employees and non-technical users, focused on modern threats, including those related to artificial intelligence. Participants will learn how modern AI-driven scams work (phishing, deepfake voice/video, fake profiles, automated scam campaigns), how to recognize them, and how to use AI tools safely at work and in everyday life. The training is practical in nature and focuses on real-world scenarios, common user mistakes, and simple rules that effectively reduce risk.
Comprehensive security assessment of your AI/ML systems, LLM integrations, MCP servers or Agentic AI. I simulate real-world attacks to uncover vulnerabilities before malicious actors do. I rely on established standards such as MITRE ATLAS and OWASP Top10 for Large Language Model Applications. You receive a detailed report with findings, risk ratings, and actionable remediation steps.
AI is already present in organizations, often outside of management’s visibility and without formal policies. I will help you regain control over AI risk without slowing business growth.
ABOUT
I combine expertise in artificial intelligence with hands-on penetration testing experience. My workshops are based on real-world attack scenarios, not theoretical slides. I have worked with technology companies and institutions across Europe.
I contributed to the development of AI security standards, including OWASP Top 10 for Machine Learning and OWASP Top 10 for Large Language Model Applications.
I participated in international security initiatives related to NATO and spoke at conferences such as BSides and Nordic AI Summit.
My articles have been published in hAI Magazine and on security-focused platforms.
I publish an AI security newsletter on hackstery.com called “The Real Threats of Artificial Intelligence.”
I am a winner of 7 AI and cybersecurity hackathons, including three HackYeah editions and the KNF hackathon. I've been a judge and a mentor at HackYeah hackathons (2024, 2025).
I hold certifications including Certified AI/ML Penetration Tester and Certificate of Cloud Security Knowledge.
Examples of platforms where I reported issues via responsible disclosure:
Logos are used for informational purposes only. No commercial affiliation.
CASE STUDIES
The financial institution needed to train its security and compliance teams on AI security topics.
An 8-hour workshop focused on AI security regulations and hands-on exercises involving attacks on chatbots and AI agents.
The workshop received a rating of 4.5/5. Participants gained practical skills in attacking chatbots and AI agents, as well as secure integration of LLMs with web applications.
The company needed to train employees as part of an AI security awareness program.
A 3-hour security awareness training covering modern AI threat classes and principles of safe AI usage.
Nearly 100 employees were trained. The workshop received very high ratings from both technical and non-technical participants, including management.
The company wanted to test the security of its AI chatbot used for product recommendations in an online store.
A security architecture review followed by targeted penetration testing, including analysis of chatbot threats described in OWASP Top 10 for Large Language Model Applications.
Several chatbot security issues were identified, including prompt injection and system prompt leakage. All vulnerabilities were remediated.
CONTACT
Looking for AI security workshops for your team or need a penetration test before deployment? Tell me about your needs and I'll respond within one business day.
contact@aifense.eu
Mikołaj Kowalczyk
Based in Bielsko-Biała · Serving Central Europe