Dr Mahault Albarracin is a leading voice in ethical AI, sustainability, and cognitive computing. As Director of Innovations & R&D at VERSES AI, she stands at the forefront of building transparent, human-centred AI ecosystems. Her research bridges science, governance, and philosophy to help shape a responsible technological future.
A highly regarded AI speaker, Mahault explores how diversity, inclusion, and ethical design can guide the evolution of intelligent systems. Her work challenges organisations to look beyond efficiency and towards accountability, long-term sustainability, and shared human benefit. She is widely recognised for making complex ideas in AI governance accessible to leaders and policymakers.
In this exclusive interview with The AI Speakers Agency, Mahault discusses the urgent need for stronger governance frameworks, the impact of AI on society and the environment, and why building ethical intelligence is the next frontier of innovation.
Q1. As AI systems become more autonomous and embedded in decision-making, how can organisations design governance frameworks that ensure transparency, accountability, and control?
Mahault Albarracin: “Impacts — I think the most pressing currently is governance, especially governance complexity, because the current governance frameworks for AI are really insufficient to manage the risks of increasingly autonomous AI systems, particularly those that we hope will be able to have full agency.
“For example, we have to factor in actor, agent, and network risks. Ethical challenges will stem from these different levels of governance, including actor governance, which focuses on regulating developers and providers; agent governance, which is about regulating autonomous agents; and ultimately, network governance, where coordination among multiple agents is going to be one of the things we’ll have to tackle — a little bit like geopolitics.
“Then we also have to care for transparency and accountability, where right now most sophisticated AI either is a black box or too complex for most people to explain the decision-making process and therefore understand and possibly justify it.
“This leads to problems of possible bias and inequality. The training data can be biased, which results in algorithmic bias in itself. It can lead to inequalities in outcomes, which are already baked into our societal systems — these are just amplified ultimately.
“But the data in itself is not the only source of bias, and identifying all these requires intense training for a lot of the people in the field. We also want to make sure, again on this point of autonomy, that we have the ability to balance the autonomy of AI — its ability to learn and make decisions — with the control that we want to have over it, a little bit like a legal system for us.
“For example, autonomous vehicles need to have this proper balance to make decisions that we are okay with them making. Then there’s all this idea about privacy violations, because there’s a lot of data mining techniques used for training, and sometimes they can be retrieved directly. There are lots of possible attacks that you can do on different models to retrieve sensitive data.
“Then there’s the energy consumption aspect of it all — this requires a lot of energy and therefore has a very high environmental impact. We have to be conscious of the carbon emissions and environmental strain of these models on the world.
“Finally, we also have to consider human design and the potential for manipulative design, where AI systems can be used to manipulate behaviour — like in advertising, or as we’ve seen in political campaigns. This ultimately leads to the possibility of coercive use and loss of informed consent.”
Q2. What practical steps can business leaders take to embed ethics, inclusion, and diversity into their technology strategies, particularly within STEM-driven industries?
Mahault Albarracin: “This is a really tricky question because ethics obviously is a tense subject. Companies have shareholders to answer to, and then they also have public perception, and ultimately, they have the perspective of the people themselves working in the company.
“There are solutions, like adopting a multi-layered governance model where you implement actor, agent, and network governance within your systems. Ultimately, you try to handle the autonomous decision-making of the AIs, manage the network governance of interactions between the agents, and use this governance framework to make sure that whatever ethical framework you adopt can be implemented within your system.
“You should prioritise transparency and trust — establish standards for transparency, ensure that all your processes are auditable, and this builds trust with consumers, stakeholders, and ultimately regulators. Make sure that processes are understandable and verifiable. This is work you can do uphill, and it leads us back to the idea of ethics by design.
“What you want to do is really incorporate the ethical considerations that you want to put in place right from the start. This implies that you will have an ethics core culture — an ethics department with power and the ability to audit all your projects, ensuring involvement from the start to the end.
“Then, there’s all the regulatory compliance. There are more and more AI regulations on the global stage — for example, the European AI Act. It focuses on risk categorisation and encourages adherence to technical standards to mitigate certain risks through certifications like the CE mark. You could follow this regulatory compliance, have your legal department pay attention to upcoming rules, and establish what you need to do in advance so that you’re not taken by surprise.
“Obviously, this requires training — you have to train your employees and stakeholders to understand issues of bias, data privacy, transparency, and ethical responsibility. This needs to be done throughout the company at every layer. You can also implement impact assessments — systems to identify and mitigate ethical risks in AI projects to ensure alignment with internal and external standards.
“This can be done either by your ethics team or automatically through available tools. You want to have diversity in your teams as much as possible — not just because of DEI standards but because they have the ability to see what others may not see, ensuring more equitable outcomes through diversity of perspective and thought.
“You also want continuous monitoring — just because you intended for something to be ethical at the start doesn’t mean it will remain that way. Things can evolve and have unexpected consequences. Finally, in your KPIs, you can include ethical AI metrics and benchmarks for ethical performance and autonomy. If you put these into your KPIs, you’ll be able to keep yourselves accountable.”
Q3. When addressing audiences, what core message do you want them to take away about the responsible and sustainable use of AI in shaping our future?
Mahault Albarracin: “I mean, again, it’s another really interesting question and a complicated answer because, as I said, AI can have environmental, societal, and economic impacts. So, it’s not just that we can use AI for good — it also, by virtue of being an actor in the world, has an impact.
“If we want it to stay positive, we can use it for the optimisation of resources — for example, developing smart cities that optimise resource allocation, helping with energy consumption, waste management, and air quality control.
“We can try to do adaptive decision-making and optimise resources in cities to lead to more efficient energy consumption. There are ways to use it for a good purpose, but you have to consider what kinds of models to use in order for the model itself not to have a large footprint.
“At Verses, we favour active inference, especially for urban sustainability, where we use a model that in itself is less energy costly. We also develop multi-agent systems that can do this resource management and focus on minimising free energy over longer periods of time.
“This is the idea of long-term environmental viability — you’re not trying to maximise one specific utility function; you’re trying to find this metastable state over long-term periods and therefore find strategies with both short-term and long-term benefits.
“Your models should also prioritise plasticity and adaptation. They shouldn’t just be systems that allow elasticity — they need to allow whatever you’re automating to adapt to new stressors and possibilities, especially as the environment changes. Everything is always moving forward, always changing, so you want to ensure resilience and flexibility over time as systems evolve.
“Static models will not be suited for this. You also need to understand that sustainability is an emergent property, so you’ll have to be able to understand and manage the interdependencies among different actors. That’s why multi-agent systems and, ultimately, network governance are going to be pretty critical.”
This exclusive interview with Mahault Albarracin was conducted by Female Motivational Speakers Agency.
SuperEditor