Predictions for 2025: Ethical AI, geospatial data and the future of digital transformation

Manish Jethwa, Chief Technology Officer at Ordnance Survey offers his predictions on ethical AI, geospatial data and the future of digital transformation.

What do you think is going to be new in 2025?

Next year, we can expect continued significant advancements in AI and machine learning, particularly with the push towards General AI.

The integration of large language models (LLMs) with more elaborate agents that can execute complex tasks on behalf of users will further reduce barriers to interaction.

In the field of geospatial, this will mean that we are able to translate natural language into precise data queries, and make geospatial datasets more accessible, mainstream, and user-friendly. These developments are likely to simplify and enhance the way in which industries approach problem-solving and decision-making.

In addition to advancements in LLMs and Generative AI, I anticipate progress in the broader category of Machine Learning (ML), driven by greater access to graphics processing units for training.

At Ordnance Survey (OS), we’ll leverage this capability to train models for specific, complex tasks such as automatic feature extraction from imagery.

With an increasing volume of data generated automatically, hopefully next year will also bring innovative tools and techniques to validate data, ensuring it can be confidently used for its intended purpose.

What’s going to stay the same in 2025?

The pace of technological advancement will remain constant, especially in AI-driven areas despite the recent slowdown in the enhancing capability of LLMs. Businesses will continue to adopt new tools to maintain competitiveness, focusing heavily on digital transformation initiatives to integrate AI and other emerging technologies.

As part of our digital transformation journey over the last 5 years, OS took the opportunity to totally redesign our data which led to the launch of the OS National Geographic Database (OS NGD). With new application programming interfaces (APIs) and a bespoke download service (OS Select+Build), we were able to address the evolving needs of our customers and ensure higher levels of usability, personalisation and rich attribution. 

However, some challenges that have historically slowed down digital transformation, such as cultural resistance and rapid successive changes leading to change fatigue, will likely persist. Companies will still need to balance the pressure to adopt new tools with the need to phase in changes carefully, ensuring that human elements like retraining and team adaptability are addressed.

Successful transformations will continue to rely on a clear vision of future goals, effective communication of progress, and celebrating milestones to sustain momentum.

Additionally, the intersection of AI adoption and secure data management will remain a critical focus area. Balancing the pressure to integrate generative AI tools with the need for responsible usage will be an ongoing challenge, which is why OS has defined its Responsible AI Charter to be mindful of the pitfalls when implementing new techniques.

What would you like to change in 2025?

I would like to see a greater emphasis on ethical AI and responsible technology development. This includes ensuring that AI systems are transparent, fair, and unbiased, and that they are developed with consideration for their environmental and societal impact.
In the context of practical applications, addressing challenges at the intersection of data infrastructure and AI investments is critical.

AI tools must be integrated responsibly into workflows, with a focus on ensuring quality, managing risks, and protecting intellectual property.

Organisations should invest in building automated workflow pipelines that incorporate confidence measures being delivered alongside the outputs. This approach ensures that consistency and reliability are easily considered and factored in prior to use.

Another change I’d like to see would be a broader commitment to retraining and upskilling employees to prepare them for the impact of AI and digital transformation. At OS this has been a cornerstone of our AI strategy, helping to mitigate risks and manage the human implications of process changes.

We need to be careful that in our aim to enhance efficiency, we don't lose the personality, creativity, and emotion that we bring as humans into the workplace. This is crucial to building strong relationships with customers and partners but also key for collaboration and teamwork that defines the business culture. Such approaches should become the norm across industries. 

ML is already playing an ever-increasing role in delivering valuable insights to organisation that have their data (both structured and unstructured) collated in a singular location. However, with cybersecurity threats being powered by AI and becoming more sophisticated, businesses will want to combat this challenge with well thought out strategies.

From where data is stored, to how it is structured and documented to allow for analysis and better decision making. This will also require a shift in the way we think about our IT functions from purely a service to a value centre.

The new wave of innovations, powered by AI, are sweeping through industries with new competitors appearing at pace. Companies that fail to keep up open themselves up to risks, such as changing customer expectation as well as attracting and retaining talent, in addition to losing their competitive edge. So, we've got to keep moving forwards.