Learning to lead AI projects: lessons from mental healthcare

The experience of a past participant of the Responsible AI Implementation Lead course
October 7, 2026

Kiki de Waal of GGZ NHN took the Responsible AI Implementation Lead course to build AI-specific expertise within her organisation. She tells us how it gave her a framework for asking the right questions, and what challenges come next.

GGZ Noord-Holland-Noord (GGZ NHN) specialises in mental healthcare for all ages across the Netherlands and has taken a leading role in shaping responsible AI adoption. This is reflected in its involvement in national networks for (generative) AI implementation, such as the RIGH:T Consortium for ambient listening. As GGZ NHN began exploring AI projects, however, it recognised the need to build up its in-house expertise to ensure these initiatives could be implemented responsibly.

There were project leaders, but no one with specific knowledge of what responsible AI implementation actually requires. Kiki de Waal, who had experience with digital tools and app implementations throughout her career, was asked to take the Responsible AI Implementation Lead course and to bring what she learned back to the AI team at GGZ NHN. She came in with an open mind and left with a structured framework for asking critical questions about her organisation's AI projects.

What led you to sign up for the course?

I was part of the AI team, and we had AI projects that needed to be implemented. GGZ NHN did have project leaders, but there was really no one who knew how to successfully implement an AI tool. I had experience implementing digital programmes, but AI tools were something new and different. Essentially no one within the organisation had specific knowledge about it, so it was decided that I would take the course to really learn that AI-specific part and bring that input back to the team.

After each course day, I'd always come back with a short list of questions: do we actually have this in place? How do we handle this in our organisation? That rhythm worked really well.

Were there specific topics where you felt you lacked a clear framework going in?

Legislation around AI, for sure. I'm not responsible for it directly, but as a project leader, someone a bit further removed from the regulations, it's valuable to know what's going on. To know what the MDR and AI Act are, when they apply and when they don't, and to follow developments as they unfold. What I also noticed is that when you're an AI project leader, colleagues start asking you questions that go deeper than what a project leader would normally be expected to know, because everything is still so new. The course gave me a broad foundation: data protection, security, a bit of everything. At a high level, but that's exactly what you need.

What did the course specifically change in how you work?

Mainly, it started conversations. After each course day, I'd come back and ask: how complete is our business case, actually? Have we thought about our AI implementation approach? Do we have the right people at the table? The course gave me  a structured way to step back and ask: what are we doing, and are we doing it well? It helped me bring more structure to the process, and flag issues before they became problems.

Did the timing and pace of the course fit well with where your project was?

Yes, very well. When I started the course, we were in the phase of just beginning conversations with AI vendors, and the course was always just one step ahead, which meant we never felt like we were running behind. The month between each session was also useful: to discuss things internally, get the right people together, and actually think something through before the next day.

I would say this course is best suited to an early stage, when you're starting out or just beginning vendor selection. If you've already implemented a tool, you may have already taken steps and maybe it would have been good to approach things differently. And if you're in a very advanced phase, the content is less directly applicable.

What stood out as most practically useful?

It's hard to name one specific thing. What I took from it most was really the whole: making sure everything is documented, that you know what you're doing and why, and that the right people are involved. The procurement tips were useful at a moment when that was exactly what we were working on. More broadly, it gave me the confidence to ask critical questions in meetings about security or legislation, for instance, where I might otherwise have felt less comfortable. I now recognise the terms and know what they're about. That matters.

Who within your organisation would benefit from a course like this?

Project leaders, first and foremost. But also advisors in learning and development who work on AI literacy, and people in IT, or information management roles. And beyond that, I think it would actually be interesting for clinicians, and perhaps even board members. AI touches so many parts of an organisation that it helps if everyone is aligned and moving in the same direction. You need everyone for that.

Looking back, did the course meet your expectations?

I went in with an open mind and no strong expectations, so I can't say it surprised me, but I can say I got a lot out of it. If there were ever another AI project leader at GGZ NHN, I would tell them to take this course. That says enough, I think.

Looking ahead, Kiki pointed to a challenge many organisations will soon face: how to successfully scale AI tools, and at what point an AI tool stops being a project and becomes part of the daily workflow. Coming from mental healthcare, she also highlighted how differently topics like consent play out in settings where patients may lack decision-making capacity, or in forensic and youth care – nuances that are especially valuable to explore with peers from the same field. We'll be taking both themes into the next editions of the course.

Kiki's experience shows what a difference it makes when someone leading AI implementation has a clear framework to work from, rather than having to figure it out through trial and error. That's exactly what the Responsible AI Implementation Lead course offers: the structure and confidence to ask the right questions from the very first conversations with vendors, well before an AI tool is in place.

Are you leading or involved in AI implementation in your organisation? Join our next cohort, starting in January 2027. We welcome participants from across healthcare, from hospitals to mental healthcare and beyond.  

Find out more about the Responsible AI Implementation Lead course here or directly contact the course lead, Sade Krijgsman, for more information and to discuss the available options.  

‍

GGZ Noord-Holland-Noord (GGZ NHN) specialises in mental healthcare for all ages across the Netherlands and has taken a leading role in shaping responsible AI adoption. This is reflected in its involvement in national networks for (generative) AI implementation, such as the RIGH:T Consortium for ambient listening. As GGZ NHN began exploring AI projects, however, it recognised the need to build up its in-house expertise to ensure these initiatives could be implemented responsibly.

There were project leaders, but no one with specific knowledge of what responsible AI implementation actually requires. Kiki de Waal, who had experience with digital tools and app implementations throughout her career, was asked to take the Responsible AI Implementation Lead course and to bring what she learned back to the AI team at GGZ NHN. She came in with an open mind and left with a structured framework for asking critical questions about her organisation's AI projects.

What led you to sign up for the course?

I was part of the AI team, and we had AI projects that needed to be implemented. GGZ NHN did have project leaders, but there was really no one who knew how to successfully implement an AI tool. I had experience implementing digital programmes, but AI tools were something new and different. Essentially no one within the organisation had specific knowledge about it, so it was decided that I would take the course to really learn that AI-specific part and bring that input back to the team.

After each course day, I'd always come back with a short list of questions: do we actually have this in place? How do we handle this in our organisation? That rhythm worked really well.

Were there specific topics where you felt you lacked a clear framework going in?

Legislation around AI, for sure. I'm not responsible for it directly, but as a project leader, someone a bit further removed from the regulations, it's valuable to know what's going on. To know what the MDR and AI Act are, when they apply and when they don't, and to follow developments as they unfold. What I also noticed is that when you're an AI project leader, colleagues start asking you questions that go deeper than what a project leader would normally be expected to know, because everything is still so new. The course gave me a broad foundation: data protection, security, a bit of everything. At a high level, but that's exactly what you need.

What did the course specifically change in how you work?

Mainly, it started conversations. After each course day, I'd come back and ask: how complete is our business case, actually? Have we thought about our AI implementation approach? Do we have the right people at the table? The course gave me  a structured way to step back and ask: what are we doing, and are we doing it well? It helped me bring more structure to the process, and flag issues before they became problems.

Did the timing and pace of the course fit well with where your project was?

Yes, very well. When I started the course, we were in the phase of just beginning conversations with AI vendors, and the course was always just one step ahead, which meant we never felt like we were running behind. The month between each session was also useful: to discuss things internally, get the right people together, and actually think something through before the next day.

I would say this course is best suited to an early stage, when you're starting out or just beginning vendor selection. If you've already implemented a tool, you may have already taken steps and maybe it would have been good to approach things differently. And if you're in a very advanced phase, the content is less directly applicable.

What stood out as most practically useful?

It's hard to name one specific thing. What I took from it most was really the whole: making sure everything is documented, that you know what you're doing and why, and that the right people are involved. The procurement tips were useful at a moment when that was exactly what we were working on. More broadly, it gave me the confidence to ask critical questions in meetings about security or legislation, for instance, where I might otherwise have felt less comfortable. I now recognise the terms and know what they're about. That matters.

Who within your organisation would benefit from a course like this?

Project leaders, first and foremost. But also advisors in learning and development who work on AI literacy, and people in IT, or information management roles. And beyond that, I think it would actually be interesting for clinicians, and perhaps even board members. AI touches so many parts of an organisation that it helps if everyone is aligned and moving in the same direction. You need everyone for that.

Looking back, did the course meet your expectations?

I went in with an open mind and no strong expectations, so I can't say it surprised me, but I can say I got a lot out of it. If there were ever another AI project leader at GGZ NHN, I would tell them to take this course. That says enough, I think.

Looking ahead, Kiki pointed to a challenge many organisations will soon face: how to successfully scale AI tools, and at what point an AI tool stops being a project and becomes part of the daily workflow. Coming from mental healthcare, she also highlighted how differently topics like consent play out in settings where patients may lack decision-making capacity, or in forensic and youth care – nuances that are especially valuable to explore with peers from the same field. We'll be taking both themes into the next editions of the course.

Kiki's experience shows what a difference it makes when someone leading AI implementation has a clear framework to work from, rather than having to figure it out through trial and error. That's exactly what the Responsible AI Implementation Lead course offers: the structure and confidence to ask the right questions from the very first conversations with vendors, well before an AI tool is in place.

Are you leading or involved in AI implementation in your organisation? Join our next cohort, starting in January 2027. We welcome participants from across healthcare, from hospitals to mental healthcare and beyond.  

Find out more about the Responsible AI Implementation Lead course here or directly contact the course lead, Sade Krijgsman, for more information and to discuss the available options.  

‍