JUAN JOSÉ LÓPEZ MURPHY
AI STRATEGY
Predictive AI uses historical data and ML to predict future outcomes. The emergence of bigger and better data sets and the use of powerful AI models to spot previously undetectable patterns is in some cases demonstrably improving our predictive capabilities, notes a recent article from the Financial Times. Well placed to lift the lid on predictive AI is Juan José López Murphy, Global Head of Data Science and AI at Globant, who has been working in data and AI for more than 18 years.
Juan José joined Globant as a data scientist in 2014. Eight years later, he is leading the team from its London office. Besides working on exciting projects and clients, he has spoken at numerous industry events and published two books about AI and data.
JUAN JOSÉ LÓPEZ MURPHY
TITLE: GLOBAL HEAD OF DATA SCIENCE AND AI
COMPANY: GLOBANT INDUSTRY: IT LOCATION: LONDON, UK
Juan José drives business model innovation at Globant by combining strategy, AI, technology and data. A passionate data evangelist, he focuses on disruptive growth while empowering his team to continuously excel.
No crystal balls, please“ First of all, understand what a prediction actually is,” Juan José says.“ If what you want is an absolute oracle telling you exactly what is going to happen, you are setting yourself up for failure.” Juan José outlines that every prediction has uncertainty and is made within a given set of parameters.
“ Every prediction only becomes useful in the context of the decisions you are going to make with it,” he adds.
Juan José highlights that leaders need to understand those bounds:“ What are we predicting? How certain or uncertain can that prediction be? What decisions are we going to make as a consequence? And, critically, how much does that uncertainty matter to the business?”
Juan José notes that, sometimes, managing the uncertainty is actually more important than squeezing another decimal point of accuracy out of the prediction itself.
“ The second thing is making sure you have something genuinely worth predicting,” he continues.“ It should be a measure that is meaningful to the business, not simply something that happens to be associated with the outcome you actually care about.”
To do that, Juan José notes, the important thing is to be explicit about what you are actually measuring, what you believe it tells you and how that connects to the decision you are trying to make.
68 October 2026