I made 2025 my AI year. From the moment I watched Sam Altman take to the stage to reveal ChatGPT, I knew my business was changed forever. I am an intentional person. I really consider where I put my effort and my energy. So when I was planning this year, I set some clear intentions.
As I write, it’s Sunday evening, the 14th December, and I reflect on the gratitude of setting those intentions and committing to the daily actions to achieve my goals.
I committed to being in the rooms where real decisions were being made, not just talking about AI in theory but seeing how it was landing in organisations, teams, classrooms, boardrooms, studios, and businesses.
I spoke on stages in Las Vegas, Spain, Brussels, and across Ireland. I sat in boardrooms with senior leaders. I trained teams in classrooms and Zoom rooms. I worked directly with clients who wanted to move from curiosity to confidence.
I launched the AI SIX podcast in February, which has now grown to almost 300 episodes. I published my fourth title, AI Playbook, in September, which sold over 500 copies and came close to breaking the top 10 in Irish book sales during launch week.
I completed the beta of my AI Diploma, which launched at the end of 2024 and ran through this year. I also did the uncomfortable but necessary work of implementing AI properly in my own business, including an AI audit, new systems, and the early build of what will become AIJ, a product I talk about in both the book and the podcast.
After 12 months of conversations, workshops, training sessions, notes, questions, resistance, excitement, and experimentation, one thing is clear to me.
2025 was the year of AI engagement.
People are no longer pretending this is optional. AI is here, it is being used, and it has entered daily working life whether organisations officially sanctioned it or not. But engagement is not the same as strategy, and it is definitely not the same as impact.
That is why 2026 has to be different.
What I actually saw in 2025
The biggest shift I noticed this year was not technical. It was psychological.
People are curious. Genuinely curious.
That might sound obvious, but it matters. Curiosity is the starting point for literacy, and literacy is the foundation for everything else. I met people who were experimenting quietly, people who were openly excited, and people who were sceptical but still paying attention. Even the most cautious organisations were no longer asking whether AI would affect them. They were asking how fast.
AI literacy is growing, largely because people are learning by doing. They are writing prompts, testing tools, breaking things, and seeing what works. That hands-on learning is essential. You cannot understand AI from a policy document alone.
But the literacy is still narrow.
Most people understand how to use a tool. Far fewer understand ethics, governance, data implications, scalability, or what happens when you move from individual use to organisational systems. There is a gap between tactical use and strategic understanding, and that gap will matter more in 2026.
Another question came up again and again.
What is the ROI of AI?
On the surface, this looks like a standard business question. But the more I heard it, the more I realised it is actually two questions pretending to be one.
What value can AI bring to me, my role, my organisation?
And what value am I bringing to AI in terms of judgment, context, experience, and responsibility?
If we only ask the first question, we reduce AI to a cost-saving exercise. If we ask both, we start to think about quality, outcomes, and impact.
Governance was the other constant theme.
There is still real nervousness about data sharing. In many organisations, tools remain locked down, restricted, or unofficial. Shadow AI is very much still a thing. People are using tools anyway, often without guidance, training, or guardrails.
That creates risk, but it also tells us something important. When people find value, they will use the tools regardless. Governance that ignores reality will always fail. Governance that supports safe, transparent, well-defined use cases has a chance of working.
Finally, there was a deeper question bubbling under the surface.
Impact.
Not just economic impact or productivity gains, but broader impact. How does AI change how we serve people, audiences, citizens, customers, and communities? How does it change how we tell stories, make decisions, design services, or build trust?
I heard this question from public sector teams, radio stations, educators, marketers, and leaders. The language differed, but the concern was the same.
What does responsible, meaningful AI use actually look like in our context?
What I learned at Harvard
Alongside the client work and training, I spent time this summer studying AI and Public Policy at Harvard. It was a chance to step back from the day-to-day and think about the bigger picture with people who are shaping policy, research, and regulation at the highest levels.
I came away with five things that have stayed with me.
First, Europe is seen as highly regulated, and the US focus is firmly on the US. This worried me. There is a real risk that Europe’s regulatory approach, while well-intentioned, could slow down our ability to innovate. I left those sessions thinking hard about what that means for Ireland, for our start-ups, for our ambitions. We cannot afford to be left behind while others move faster.
Second, domain expertise gives us an edge. AI is powerful, but it is not a substitute for knowing your field. The people who will thrive are the ones who combine deep knowledge of their industry with an understanding of what AI can and cannot do. That is where real value gets created.
Third, ethical AI requires an all-of-society approach. Legislation alone will not get us there. We need educators, businesses, civil society, and citizens all engaged in the conversation. This is not something governments can solve in isolation. It requires everyone.
Fourth, we have more questions than answers. And that is okay. Anyone who claims certainty about where AI is heading is either selling something or not paying attention. The honest position is to acknowledge the uncertainty and keep learning.
Fifth, I was right to follow the ‘AI for Good’ mantra. Hearing from researchers, policymakers, and practitioners reinforced what I have believed for years. The goal is not AI for its own sake. The goal is AI that serves people, improves outcomes, and does more good than harm.
There was a sixth thing too, though it is harder to articulate. Being in those rooms gave me validation. Validation that what I am building, the frameworks I use, the approach I take with clients, it holds up. That mattered more than I expected.
From engagement to impact
Engagement is necessary. Without it, nothing moves. But engagement alone is not enough.
Engagement is trying out tools. Impact is changing outcomes.
Engagement is experimentation. Impact is intention.
Engagement is individuals figuring things out. Impact is organisations aligning around purpose, policy, and practice.
If 2025 was the year we collectively stopped ignoring AI, then 2026 must be the year we stop using it casually.
That means moving from scattered use to strategic implementation.
It means defining clear use cases that are tied to real problems, not novelty. It means understanding where AI adds value and where it does not. It means investing in literacy beyond prompt writing, including ethics, governance, and decision-making. It means designing systems, not just workflows.
It also means being honest about impact.
AI will not magically fix broken processes. It will not compensate for a poor strategy. It will not replace leadership, judgment, or accountability. Used badly, it will amplify existing problems. Used well, it can free up time, improve quality, and help organisations focus on what actually matters.
That is the work ahead.
What I am focused on heading into 2026
As I look to the year ahead, my own focus mirrors what I believe organisations need to do.
The AI Playbook Diploma launches in January, bringing together everything I have learned from the original diploma, the book, client work, and this year’s training. It is designed to be current, practical, and grounded in real-world use, not theory.
My AIJ product will enter beta in Q1 2026, with a small cohort trialling it in practice. This is about systems, not shortcuts. About embedding AI in ways that support decision-making rather than replacing it. If you want to be one of the first organisations to use it, please get in touch.
And in my own business, the work continues. AI audits, system design, governance decisions, and constant reassessment. If I am asking organisations to do this work, I have to be willing to do it myself.
After a year of listening, teaching, building, and testing, my biggest takeaway is simple.
If 2025 was the year of AI engagement, 2026 has to be the year of AI impact.
Not louder conversations. Better ones.
Not more tools. Better use.
Not faster adoption. Smarter implementation.
AI is no longer a future problem. It is a present responsibility. What we do next will matter far more than what we experimented with this year.
A big thanks to everyone who worked with me this year, to Harvard for accepting my application back in March, to my team, who are as excited as me to show up for work every day, and to my clients, who really motivate me to be better so that you can deliver better.
I want to take this opportunity to wish you all a great break, and here’s to more AI for Good in 2026.
About the Author
AI and Communications Specialist | Author | Podcaster | International Speaker
Joanne Sweeney is a recognised leader in digital marketing and digital transformation, specialising in the practical application of AI to enhance communication strategies. The founder of Digital Training Institute, Joanne empowers leaders and communications professionals to leverage AI responsibly and effectively, ensuring transparency, trust, and tangible results.
She completed the Harvard Kennedy School Executive Programme in AI and Public Policy this summer before launching her fourth book AI Playbook, in September. Joanne has also published two editions of Public Sector Marketing Pro, which were preceded by her first title, Social Media Under Investigation Law Enforcement and the Social Web.
Her books have become go-to resources for governments, public sector and policing agencies, as well as marketing and communications professionals, all navigating the challenges of the changing Internet.
Joanne is the host of the AI SIX Podcast, a daily six-minute podcast which provides actionable insights, expert interviews, and practical advice to help professionals navigate the new world of work.
