Red-team your own chatbot or model
Test a system you run or plan to run for prompt injection, data leakage, bias and failure modes.
Design AI that is safe, fair and secure. Your team tests the systems you run or plan to run for bias, robustness, adversarial attacks and compliance.
Plan this course for your team
£68.75 per person after funding
Get a funded in-house quote Apply for funding on Business WalesIndicative, for eligible Wales-based businesses. You pay the course fees and the Welsh Government reimburses its share after the training. Funding is discretionary and capped at £50,000 per application.
Public courses run in Bridgend at £275 per person.
New public dates are being scheduled. Request a place and we'll tell you the next date, or book a private course for your team.
Request a public place
Flexible Skills Programme · Welsh Government
The Flexible Skills Programme's re-launched AI Skills Pilot runs until the end of March 2027 and focuses on the next stage of AI learning: business-specific use cases and upskilling. That is exactly how we run in-house courses. Eligible Wales-based businesses can claim back 75% of approved training costs, up to £50,000 per application.
Relevant Business Wales training area: AI-specific skills: AI ethics, AI deployment and integration. Information and cyber security.
151days left to complete funded training under the AI Skills Pilot (ends 31 March 2027)
Funding is discretionary and the final decision on eligibility and award rests with the Welsh Government. Apex Training Wales is a training provider and does not decide eligibility. Source: Business Wales, Flexible Skills Programme and AI Skills Pilot update (October 2026).
This course equips learners with the knowledge and practical skills to design AI systems that are safe, secure, and ethically responsible. Participants explore issues such as algorithmic bias, fairness, robustness, adversarial attacks, and regulatory compliance. The course combines theory and hands-on projects, enabling learners to implement security measures, evaluate system risks, and ensure their AI models are reliable and trustworthy. By the end, participants will understand best practices for building AI systems that are safe, ethical, and aligned with industry standards.
This course is ideal for AI developers, data scientists, software engineers, and technical professionals responsible for designing or deploying AI systems. It is also suitable for managers, project leads, and regulators who want to understand ethical and secure AI practices.
Learners begin by exploring key concepts in AI safety and ethics, including bias, fairness, transparency, and accountability. The module covers regulatory frameworks, ethical AI guidelines, and real-world case studies, helping participants understand how to design AI systems responsibly and avoid common pitfalls.
This module focuses on building AI models that are resilient to attacks and failures. Learners explore adversarial examples, model robustness, secure data handling, and threat mitigation strategies. Hands-on exercises involve testing and strengthening models to ensure reliability and security in critical applications.
Learners apply ethical and security principles to real-world AI projects. The module covers model auditing, fairness evaluation, secure deployment, and compliance with industry standards and regulations. Participants gain the skills to design AI systems that are not only effective but also trustworthy, safe, and ethically aligned.
Attending this course ensures learners can create AI systems that are safe, reliable, and ethically responsible. Participants gain practical experience in risk mitigation, bias reduction, and compliance, preparing them to lead the development of trustworthy AI solutions in critical industries.
This course gave me a deep understanding of AI safety and ethics. The practical exercises on bias detection, adversarial attacks, and secure deployment were invaluable. I now feel confident designing AI systems that are responsible and reliable.
In-house, built around your business
Generic online courses teach the tool. We bring AI Safety and Ethics to your workplace and build the exercises around the jobs your team actually does, so the skills are in use the next morning. Business-specific use cases are exactly what the refreshed AI Skills Pilot is designed to fund.

Test a system you run or plan to run for prompt injection, data leakage, bias and failure modes.
Turn your AI policy into concrete technical and process controls that someone owns.
Model cards, risk registers and test records in the form auditors and customers expect to see.
A short call is enough for us to recommend a course or put together a programme and a quote.