Most AI systems are trained once and deployed into a world that keeps changing. For over twenty years, my work has been about closing that gap: machine learning that learns continuously, detects when the world has drifted, and keeps performing in production.
I am a Research Professor and Founding Director of Te Ipu o te Mahara — the AI Institute at the University of Waikato, and a Professor at Télécom Paris (Institut Polytechnique de Paris), where I serve as Co-Scientific Director of the FinAI-Lab, the joint research laboratory founded by BNP Paribas and Télécom Paris to build trustworthy AI for critical, highly regulated financial environments.
I have been involved in several of the open-source foundations of stream learning — MOA (Massive Online Analysis), River, CapyMOA, and most recently TuiML, machine learning for AI agents — and I co-authored Machine Learning for Data Streams (MIT Press). These tools run today in banks, telcos, and environmental monitoring systems worldwide. I am a Fellow of Engineering New Zealand, received a Best Paper Award at ACM FAccT 2023, and served as General Co-Chair of ECML-PKDD 2024.
Industry & application focus: I partner with financial institutions, telcos, startups, and environmental organisations to take adaptive AI from pilot to production — concept drift, real-time monitoring, low-latency inference, fraud and anomaly detection, and end-to-end MLOps. Through TAIAO, New Zealand’s MBIE-funded environmental data science programme, my team applies real-time machine learning to climate and ecosystem data, from flood prediction to species detection.
Leadership & community: I co-chair the NZ AI Researchers Association (500+ members) and authored its whitepapers on Aotearoa’s strategic approach to AI; serve on the Executive Committee of the AI Forum New Zealand; contribute to international AI standards as a member of the NZ delegation to ISO/IEC SC 42; and advise internationally on AI governance. I have supervised 15+ PhD completions and champion open source as the engine of trustworthy, reproducible AI.
How can I help:
- Advise boards and executives on AI strategy, governance, and responsible AI
- Design adaptive ML systems for streaming and sensor data, drift, and real-time decisions
- Structure industry–research partnerships — joint labs, R&D programmes, tech transfer
- Deliver keynotes, masterclasses, and executive briefings
- Shape open-source strategy: licensing, community, sustainability
- Mentor and build high-calibre research and engineering teams
What’s new
- Co-Scientific Director of the FinAI-Lab, the joint AI research laboratory of BNP Paribas and Télécom Paris
- Launched TuiML — open-source machine learning for AI agents, developed at the AI Institute, University of Waikato; presented in an NVIDIA webinar on secure agentic AI
- Keynote Speaker, Australasian Joint Conference on Artificial Intelligence 2026
- 2 papers accepted at AAAI-26 Main Track, 1 paper accepted at AAAI-26 Journal Track
- Elliker, C., Read, J., Vanier, S., Bifet, A. (2026). Simulation-Driven Railway Delay Prediction: An Imitation Learning Approach. AAAI Conference on Artificial Intelligence (AAAI).
- Yu, P., Chen, Y., Xu, C., Bifet, A., Read, J. (2026). Binary Split Categorical Feature with Mean Absolute Error Criteria in CART. AAAI Conference on Artificial Intelligence (AAAI).
- Marcos Vinicius Ferreira, Matheus Souza, Tatiane N. Rios, Islame F. C. Fernandes, Jorge Nery, João Gama, Albert Bifet, Ricardo Araújo Rios(2026). Salvador Urban Network Transportation (SUNT): A Landmark Spatiotemporal Dataset for Public Transportation. AAAI 2026: 39864
- Organizing AI for Environment Science Workshop at AAAI-26
- Keynote speaker at Streaming Continual Learning Bridge at AAAI-26
- 3 paper accepted at KDD-26
- Mungari, S., Bifet, A., Manco, G., Pfahringer, B. (2026). ARES: Anomaly Recognition Model for Edge Streams. In Proceedings of the 32nd ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD).
- Mezhoudi, N., Lachaud, G., Diao, Y., Möller, T., Barry, M., Bifet, A. (2026). DyGADBench: A Comprehensive Benchmark for Anomaly Detection in Dynamic Graphs. ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), Datasets and Benchmarks Track.
- Cerqueira, V., Gomes, H. M., Heyden, M., Pfahringer, B., Bifet, A. (2026). A Framework for Evaluating and Benchmarking Concept Drift Detection Methods. ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD).