Tag: machine learning

  • AIB Debuts AI-Driven Mobile App to Deliver Personalized Financial Insights

    AIB Debuts AI-Driven Mobile App to Deliver Personalized Financial Insights

    Allied Irish Banks (AIB) has launched a completely redesigned mobile banking application, marking what the lender calls its most significant digital channel update in over a decade. The new app, rolling out from late June, leverages machine learning and advanced data analytics to turn everyday transaction data into actionable, personalized financial guidance.

    The upgrade addresses a key industry challenge identified by AIB’s own research: while 76% of Irish adults check their banking app multiple times a week, 47% rarely use it for financial insights. The app aims to close this gap by embedding AI-powered intelligence directly into the user experience, moving beyond basic balance checks and payments.

    Developed over 18 months with extensive customer collaboration and pilot programs, the app introduces intelligent spending categorization, merchant-level analysis, and proactive budget recommendations. Machine learning algorithms analyze transaction histories to surface spending trends, helping users set goals and make more informed decisions. This directly responds to the 31% of consumers who report low confidence in managing their finances and the 24% who find long-term planning frustrating.

    Security remains a cornerstone of the new platform. It integrates passkey authentication, intelligent card controls (including freeze functionality), and what AIB describes as best-in-class cyber security technology — likely leveraging ML for fraud detection and anomaly identification. The bank emphasizes that trust is foundational, supported by secure, resilient technology.

    The app is built on a cloud-based, modular architecture designed for continuous delivery and iterative improvement. This platform approach enables faster deployment of new AI features and positions AIB to compete against a fragmented fintech ecosystem. Upcoming enhancements include tools for children and parents, goal-based savings pots with predictive modeling, and mortgage management with personalized recommendations.

    AIB maintains a hybrid service model with 170 branches and ongoing investment in human support, reflecting that 94% of customers still value access to human assistance. Geraldine Casey, Managing Director of Retail Banking, stated: "This new AIB app is a major step forward in digital innovation and security for our customers, providing the convenience and accessibility of best-in-class banking they can trust." Chief Operating Officer Graham Fagan added: "We’ve built a digital engagement platform that sits beneath the app that is designed to enable us to continuously add to it quarter on quarter."

  • Exploring MIT’s Latest Machine Learning Breakthroughs in Robotics and AI

    Exploring MIT’s Latest Machine Learning Breakthroughs in Robotics and AI

    MIT continues to push the boundaries of machine learning with a series of groundbreaking research developments that span robotics, AI efficiency, material science, and more. Recent projects highlight the institute’s commitment to advancing both theory and practical applications.

    In robotics, researchers have developed a system that leverages large language models to help robots interpret vague instructions and focus on crucial details, improving task performance in homes and factories. Another innovation, known as Murakkab, optimizes multistep AI workflows, enhancing speed and energy efficiency. A new low-power chip enables tiny robots to generate 3D maps for navigation with minimal memory and power consumption.

    Beyond robotics, MIT scientists are modeling metal alloys at atomic scales to predict material properties more accurately, while game theory research shows that generalist algorithms can outperform specialists in certain scenarios. A novel spatial memory system allows robots to efficiently remember object locations, and a major update to random utility models—dubbed ‘the power of three’—improves preference prediction accuracy.

    Commercial applications include a startup using MIT technology for real-time product tracking in retail, manufacturing, and logistics. The NSF has renewed support for the MIT-led Institute for Artificial Intelligence and Fundamental Interactions (IAIFI), expanding its second phase with increased funding and broader ambitions. Researchers are also teaching AI agents to ask better questions using the game Battleship, and a new dataset called ChartNet enhances vision-language models’ ability to interpret charts.

    Additional milestones include MIT economist Whitney Newey receiving the Erwin Plein Nemmers Prize, new AI chemistry models designed by Connor Coley, and the appointment of Justin Solomon as associate dean of engineering education. MIT Open Learning launched a universal AI education program, making AI fluency accessible worldwide.

    These achievements underscore MIT’s role as a leader in machine learning, driving innovations that shape the future of technology and science.