Panel session: AI for Photonics
Wednesday, February 25th 2026, 1100-1230
This panel will discuss where AI can be a genuine help to network operators in the various challenges associated with their networks in future. Main scope will be optical networks, but we can broaden to include access, RAN and DCI where appropriate. Areas of interest will include:
- Network planning – making use of AI tools to improve the complex process of network design and equipment placement
- Network management and operations – in an era where network operators are losing in-depth skills, can AI make the overall task of network management simpler. Could discuss intent-based etc. One example would be around provisioning of new services rapidly / more dynamically.
- Network maintenance – interpretation of vast amounts of data to yield pre-emptive warnings of potential faults etc and improve overall network reliability
- Operator data becomes a precious commodity for training models and potentially vendor AI tools could benefit from multiple data sets – but how would this work from a data confidentiality perspective
- AI is not only NOT 100% reliable, but when it goes wrong, it can be dramatic, and without reason. How will operators build confidence in it and take advantage of the obvious benefits?
Participants:
- Chair: Alejandra Beghelli, BT
- Panelists:
– Jorg Peter Elbers, VP Advanced Technology, Standards and IPR, Adtran
– Nigel Davis, Systems Design Architect, Ciena
– Annalisa Morea, Senior Consulting Engineer, Nokia
– Lareb Zar Khan, Optical Network Engineer – Meta
– Michael Doherty, PhD student, University College London
Alejandra Beghelli
Alejandra Beghelli is a Research Scientist at the Optics & Quantum Centre of Excellence at BT Group, where her research focuses on future optical and quantum networks. She also holds an Honorary Associate Professor position at the Electronic & Electrical Engineering Department at UCL.
Nigel Davis
Nigel Davis is a Systems Design Architect at Ciena and a key innovator in multi-layer information modeling. He was recognized as a Ciena Distinguished Engineer in 2014 and Ciena Fellow in 2015. He was highly active in the TeleManagement Forum, where is recognized as a Distinguished Fellow. Nigel had major involvement in Open Networking Foundation where he formed and co-led the Open Information Model & Tooling project and was the editor of and key contributor to the Core Information Model (reproduced as ITU-T G.7711). Within Ciena Nigel was deeply involved in development of the Self Optimizing Fabric architecture pattern and has a long-term ongoing engagement in Management-Control product architecture working closely with the design teams. Nigel is currently the editor of and a key contributor to the Linux Foundation Transport API specifications adopted by many operators as the interface for optical network controllers. He is engaged in development of several IETF drafts aiming for advancements in the modeling of control solutions and for convergence of control models and architectures. As part of the work in IETF Nigel is developing a method for representation of component capabilities in the context of AI Ops and digital twins. Nigel has many patents on management and control.
Michael Doherty
Michael is a PhD student in the Optical Networks Group at UCL, researching reinforcement learning for optimized operation of fibre optic networks. In 2025, Michael was an Enrichment Scholar at The Alan Turing Institute and worked on AI for quantum circuit design at Quantinuum. His research has been presented at ONG, ECOC, and ONDM conferences. Michael holds an MSci in Physics from Imperial College London.
Jörg-Peter Elbers
Jörg-Peter Elbers is VP of Advanced Technology, Standards & IPR at Atran, responsible for global technology strategy, applied research, standardization, and intellectual property. He has over 25 years of experience in the telecommunications industry, with previous technical leadership roles at ADVA, Ericsson, Marconi, and Siemens. Jörg holds a Dr.-Ing. (PhD) degree in Electrical Engineering from the Technical University of Dortmund, Germany.
Annalisa Morea
Annalisa Morea is a telecommunications expert with over two decades of experience in optical networking and software architecture. Since 2024, she has been serving as a Consulting Engineer in Nokia’s Pre-Sales team, where she provides technical leadership and strategic support to customers, helping design and optimize advanced optical transport solutions tailored to business needs. Prior to this role, she worked as a Software Architect/Engineer at Nokia (2015–2023), leading the development of multilayer network planning tools, with a focus on optical line design, OTN/WDM integration, and GMPLS-based algorithms. Earlier in her career, she was a Research Engineer at Bell Labs, contributing to next-generation optical network architectures, routing algorithms, and collaborative European projects. Annalisa holds a Ph.D. in Computer Science and Networks from Télécom Paris and France Telecom R&D, and an Engineering Diploma in Electronics from Politecnico di Torino and Télécom Paris. She has authored over 50 scientific publications, filed 20 patents, and served as Associate Editor for OSA JOCN and member of several Internaltional telecommunication conferences (e..g., OFC). Her achievements include the Nokia DMTS recognition (2017) and the Fabio Neri Best Paper Award (2016).
Lareb Zar Khan
Lareb Zar Khan is an Optical Network Engineer at Meta. He received his PhD from Scuola Superiore Sant’Anna in Italy, where his research focused on data quality for training machine learning models and the computational complexity of models used for Failure Management of Optical Networks. He also worked at Infinera (Portugal) and Orange (France) for 18 months during his PhD.