Dr Sheelagh Brady

AI for Good: Kowroo App uses AI to make travel safer

Submitted on Wednesday, 05/06/2024

Former police officer and Insight member Dr Sheelagh Brady discusses her reasons for developing an AI-powered app to keep people safe when they travel, in Silicon Republic

Dr Sheelagh Brady has 25 years’ experience in policing and security. She has worked with An Garda Síochána, the EU Common Security and Defence Policy assignments and the UN on various international missions.

For the last 10 years, she has consulted with businesses, governments and NGOs on security and risk management.

Brady obtained a PhD in criminology and terrorism studies from Dublin City University, where she is now developing a new travel safety app called Kowroo. She is the commercial lead for this AI-powered platform which aims to support travellers by providing security and safety alerts in real time. Researchers at the Insight SFI Research Centre for Data Analytics are supporting the development of Kowroo, which is funded by an Enterprise Ireland commercialisation grant.

Tell us about your current research.

Evacuation from Libya culminated in myself and my co-founder, John Roberts, having the idea for Kowroo. We could see first-hand that there was a need to optimise the increasing data flows to produce reliable, uncluttered information that was timely and actionable for people travelling.

We have honed the idea over years, through conversations with security managers, business travellers, technologists and many others, and in January 2022 we started the journey to bring this to life. By bringing together a team that combines tech and business insights, we have developed Kowroo, which uses the latest AI to leverage the power of multiple global datasets to produce personalised actionable insights in real time to travellers and support teams.

In your opinion, why is your research important?

Kowroo addresses a number of problems that many companies face. It enables them to comply with their duty-of-care obligations to their employees in a manner that users engage with, addressing the current issue of underutilisation of travel risk management solutions.

It does this by providing greater insights into the threat landscape for both the company and travellers. Companies tend to concentrate on incidents that have the potential for significant impact, whereas employees are far more concerned with incidents that directly impact them, such as sexual assault, theft, medical issues etc. In fact, over 60pc of business travellers have had or know someone who has had a negative experience while travelling for business, yet the majority are unreported. This statistic rises to 80pc for females and members of the LGBTQ+ community. If these incidents are not reported, there is a significant knowledge gap in any threat analysis.

The implications of this are considerable, negatively impacting the wellbeing of travellers, their productivity and frequency of travel, with some reporting not applying for positions that necessitate business travel. In an era where retention and recruitment can be challenging, this can have a considerable impact on companies.

What inspired you to become a researcher?

Coming from the operational field, I always felt things could be done better. A chance encounter while studying at the John Jay College of Criminal Justice in New York with criminologist David Kennedy made me realise there was a need for people to research and innovate at the intersection of academia and operations, and since then, I found my niche.

I try to bridge the gap by applying academic rigour and understanding to problems often viewed as impossible to solve, due to incomplete or contradictory data or as a result of an ever-changing landscape. This is very much the case in the security field, which is further hindered by an often-traditional perspective on the way things should be done.

What are some of the biggest challenges or misconceptions you face as a researcher in your field?

The biggest challenge I have faced is communicating my research in a manner that has relevance and is relatable in the operational setting. Now developing Kowroo, the challenge is ensuring the technical developments are not only innovative, but also meet a current commercial need. This is very apt at the moment, as we start trialling Kowroo in the live environment to test the next phase of market fit. It’s both an exciting and challenging position to be in.

Do you think public engagement with science has changed in recent years?

In the context of the often traditional arena I tend to work, public engagement – like technical innovation – has been slow, but this is changing at speed, especially with AI and machine learning opportunities.

The concern or challenge going forward, however, is that developments meet a real operational need, which is not always the case. Many promises have been made in this sector, yet have underdelivered or worse still, have caused additional problems, which has eroded trust in the ability of technology to address problems in this space.

This has made us very focused on the process behind how Kowroo works, including structures for checks and balances by keeping humans in the loops as we develop our modelling.

Security expert develops AI-powered app for safer travel