Recorded live at Stendhal Festival, Northern Ireland
When people talk about Artificial Intelligence, the conversation often swings between two extremes. On one side are those who see AI as the solution to almost every problem. On the other are those who view it as an existential threat. The truth, as is often the case, lies somewhere in between.
That was the focus of our recent Stage the Change event at Stendhal Festival, where we brought together experts from very different disciplines to explore AI from multiple perspectives. Rather than debating whether AI is good or bad, we asked a more important question:
How do we use AI responsibly?
Stage the Change was created to give changemakers a platform to challenge perspectives, share ideas, and spark meaningful conversations. There could hardly be a better venue than Stendhal Festival, where creativity, culture, community, and innovation come together.
For this discussion, four speakers brought unique insights:
- Virginia Mendez examined AI through the lens of gender equality and social justice.
- Esteban Rocha explored the opportunities AI creates for innovation and problem-solving.
- Allistair Wilkinson revealed the hidden infrastructure powering the AI revolution.
- Izy Boyd discussed the environmental impact of AI and the need for greater transparency.
The chair for the discussion panel: Vincent Lusby, a researcher at Queen’s University Belfast, specialising in decision-making, governance and sustainable development. His work focuses on how systems, policies and institutions can deliver better outcomes for people and the environment, bringing a valuable perspective to discussions around the societal impact and accountability of artificial intelligence.
Together, their perspectives painted a picture that was far richer and more complex than the headlines often suggest.
AI Is Not Neutral
One of the first challenges raised during the session was the assumption that AI is inherently objective.
Virginia Mendez highlighted how AI systems learn from existing human behaviour and historical data. When the data contains bias—as society often does—AI can amplify and scale those inequalities. Examples ranged from recruitment systems that disadvantaged women to criminal justice tools that disproportionately impacted racialised communities.
Her key message was simple but powerful:
AI does not exist outside society. It reflects the values, priorities, and assumptions that we build into it.
This means fairness is not automatic. It is a choice. Diversity, inclusion, and ethical design cannot be afterthoughts—they must be part of the development process from the beginning.
The Extraordinary Potential of AI
While the discussion explored risks, it was equally important to recognise the opportunities.
Technology specialist Esteban Rocha described how AI is enabling what he called a “Cambrian explosion” of applications. Tasks that previously required significant funding, technical expertise, and months of development can now be achieved in a fraction of the time.
He shared a particularly moving example from Venezuela, where communities affected by an earthquake rapidly created AI-assisted tools to locate missing people, map damaged buildings, and coordinate support efforts. In situations where every minute matters, AI dramatically reduced the barriers to creating practical solutions.The lesson was clear: AI is an incredibly powerful tool. Used responsibly, it has the potential to empower communities, enhance creativity, and accelerate innovation at unprecedented speed.
However, with that power comes responsibility.
The Infrastructure We Rarely See
One of the most fascinating perspectives came from materials scientist and cement chemist Allistair Wilkinson.
When people think about AI, they often imagine software, algorithms, and digital assistants. What they rarely think about is concrete.
Yet behind every AI prompt is a vast physical infrastructure of data centres, energy systems, cooling technologies, steel, silicon, water, and construction materials. AI may feel weightless, but its environmental footprint is very real.
Allistair highlighted the scale of modern data centres and challenged us to think about AI not simply as a technological revolution, but as an infrastructure revolution. Every interaction with an AI model relies on physical systems that require resources to build and operate.
The challenge facing industry is therefore twofold:
- Continue advancing AI capabilities.
- Build the supporting infrastructure in a more sustainable and circular way.
The Hidden Impact of Every Prompt
During my own contribution to the discussion, I introduced the concept of the Digital Visibility Gap.
When we type a question into ChatGPT, Gemini, Claude, or another AI platform, we see the answer appear almost instantly. What we don’t see are the resources consumed behind the scenes. The cooling systems, energy demands, data processing, and infrastructure all remain largely invisible to the user.
This invisibility creates a challenge.
Most people understand the impact of the fuel they put into a car because they can see the cost. We understand the energy efficiency of household appliances because labels provide transparency.
What if AI worked the same way?
What if every interaction came with information about energy use, carbon impact, or water consumption?
The goal isn’t to make people feel guilty for using AI. The goal is to create awareness and encourage informed choices. Transparency helps consumers, businesses, and policymakers make better decisions.Who Is Responsible?
One of the most engaging parts of the event was the audience discussion around accountability.
Who is responsible when AI gets something wrong?
The developers?
The technology companies?
Governments?
Or the users themselves?
The panel broadly agreed that responsibility is shared. Technology companies need transparency. Governments need effective regulation and governance. Educational institutions need to build critical thinking skills. And users need to engage with AI thoughtfully and responsibly.
As Virginia Mendez pointed out, we must also follow the money. The incentives shaping AI development matter. The decisions made by governments, regulators, investors, and large technology companies will significantly influence the future direction of AI.
Education Is the Missing Piece
A recurring theme throughout the discussion was education.
AI literacy can no longer be viewed as an optional skill. Understanding how these systems work, where their limitations lie, and how they can be used responsibly is becoming increasingly important for people of all ages.
The challenge isn’t simply teaching people how to use AI.
It’s teaching people how to question it.
How to challenge it.
How to verify information.
And how to recognise both its opportunities and its risks.
The Next Frontier
As the session drew to a close and the festival music began competing with our microphones, we ended with a simple reflection:
The next frontier of AI is no longer intelligence. It is accountability.
The technology will continue to improve. Models will become faster, more capable, and more accessible.
The bigger question is whether our systems of governance, education, sustainability, and ethics can keep pace.
AI is here to stay.
The challenge now is ensuring that the future we build with it is one that benefits people, communities, and the planet.
And perhaps the next time you open ChatGPT, Gemini, Claude, or any other AI tool, you’ll pause for just a moment and ask not only:
“What can AI do?”
But also:
“What impact does this have, and what responsibility do I have in how I use it?”
No responses yet