Rapid shifts in financial technology call for sharper judgment, stronger ethics, and informed decision making. As Dr. Janet Bastiman shares her perspective, you are invited to explore how innovation, regulation, and human insight work together to shape a more transparent and resilient financial landscape.
Your career spans both academia and industry, from computational neuroscience to leading data science initiatives. What inspired this transition, and how has your scientific background influenced your approach to applied AI?
My PHD was considered niche at the time and there weren’t many post-doctoral opportunities to apply my trade, so I turned to industry to make use of my computing skills. That ended up being a longer-term plan than intended as I worked my way from a developer up to a CTO.
When commercial AI rose to prominence in 2015, I had all the skills to build out AI functions for businesses using my practical implementation background across multiple industries.
This has led to a productionisation approach, taking AI from theory to practice.
As Chief Data Scientist at Napier AI, you operate at the intersection of technology, regulation, and financial services. What aspects of your role do you find most impactful in shaping the company’s innovation strategy?
It’s about the art of the possible balanced with the wits of what’s allowed and what’s ethical.
There can often be competing business priorities, but innovation should be rooted in what’s right beyond anything else. When companies use Napier AI’s systems, they can be confident of the impact and confident in the ethics behind it.
It’s about the art of the possible balanced with the wits of what’s allowed and what’s ethical.
Especially when tackling financial crime, which the Napier AI / AML Index 2025 – 2026 outlines as $5.5 trillion drain on global economies, having a broad understanding of global regulations, inbuilt morals and ethics, and where research is going is vital to tackle this global challenge innovatively.
You have guided teams through significant transformation—moving from traditional development frameworks to agile, cloud-based systems. What have been the most important lessons in managing people and technology through such change?
Education and communication. It’s natural human instinct to fight change, so having change management and giving people a clear reason for what you’re making a change is vital. Make sure the team understand why a change is happening, challenges and solutions, and if the team can be involved in the decision making that’s even better.
That helps get everyone on the same page and have the whole team working towards the same goal. Having people on your side makes everything so much easier.
The financial sector has embraced AI for everything from compliance to fraud detection. In your view, what are the most promising developments in AI for finance, and where do you see the greatest untapped potential?
The untapped potential is in taking a step back and looking at what pain points you have, and where AI is really the answer. There may be other tools that provide greater value and avoid running long transition projects that don’t end up delivering what’s needed, similarly trying to solve a problem too simplistically can cause false confidence in incorrect results.
For financial services, the most promising developments are tackling evolving behaviours in financial crime. AI can trace behaviours across networks and identify unusual patterns versus people who have different financial setups. This can help with false positive reduction and detecting unusual signals through the noise. $3.3 trillion could be returned to global economies with AI-powered anti-money laundering strategies.
The key is summarising this information and documentation with clear explainability.
Napier AI operates in a space where innovation and regulatory compliance must coexist. How do you ensure that new technologies enhance efficiency without compromising ethical and security standards?
It’s all about guardrails.
We don’t feed it information about individuals’ protected characteristics because that’s irrelevant to the output. We want it to learn purely from behaviours, so restricting behaviours is critical.
Whenever Napier AI look at creating new models, whether it’s traditional or generative AI, we start with the purpose of the model. When it’s a model looking at unusual transaction behaviour, for example, we don’t feed it information about individuals’ protected characteristics because that’s irrelevant to the output. We want it to learn purely from behaviours, so restricting behaviours is critical.
It’s also important to have a team of diverse thinkers from different backgrounds to provide wider awareness of socio-economic and cultural differences that can have a significant impact in areas such as financial analysis.
Collaboration between data scientists, engineers, and executives is crucial for meaningful progress. How do you foster a shared understanding of AI’s capabilities and limitations across such diverse teams?
I focus on building out teams of people who are good communicators. If you can’t convince the business that certain solutions or specific algorithms are good, then they’ll never go into production. That requires people who can talk clearly about work to stakeholders that don’t have a maths or data science background.
As AI continues to evolve, how do you envision its role in redefining the global financial ecosystem, particularly in improving transparency, risk management, and decision-making?
The number of transactions is growing year-on-year, and financial crime is growing, requiring AI’s intervention to help handle the vast volumes of data.
The whole finance ecosystem is also changing rapidly, so data that is being used to train models goes out of date quickly, and this provides challenges to systems built on historic data. With this, AI tools can help with risk management to create a safer environment for society and ensure criminals aren’t exploiting transactions as the intersection between culture and money evolves.
But this must come with transparency to make sure financial institutions understand how decisions were made.
Looking ahead, what skills or mindsets do you believe will be essential for the next generation of data leaders navigating the increasingly complex world of finance and technology?
Open curiosity and diversity of scientific thought. People need to understand how to form and test a hypothesis, as well as determine whether data provided is good data.
Combine that with core statistical understanding to understand margins of error and the context of data points. When it comes to financial crime, a few percentage points can be the difference of millions.
Executive Profile
Chair of the Royal Statistical Society’s Data Science and AI Section and member of FCA’s Synthetic Data Expert Group, Janet started coding in 1984 and discovered a passion for technology. Janet has spent over two decades helping both start-ups and established businesses implement and improve their AI offering prior to applying her expertise as Chief Data Scientist at Napier AI.
