The future of AI is built on purpose, trust and interoperability. Fabio Terezinho tells us why

The future of AI is built on purpose, trust and interoperability. Fabio Terezinho tells us why

The next wave of transformation will be powered by practical AI, secure cloud solutions and strong partnerships, says AVEVA’s vice president of integration and technology partners 

By Laura Hyde |


System integrators from across Europe gathered in Berlin, Germany, in June for AVEVA ENABLE EMEA 2025, to discuss the latest advancements in AVEVA’s industrial software portfolio. The three-day programme combined technical sessions, hands-on labs and in-depth discussions on key topics such as the CONNECT platform and AVEVA PI System and Operations Control, giving participants a chance to explore how these technologies can be used to drive innovation and efficiency in industrial operations. 

To delve deeper into the major themes of the event and learn more about AVEVA’s ongoing partnership with Microsoft, we spoke with Fabio Terezinho, vice president of integration and technology partners at AVEVA. 

You met with senior Microsoft executives at AVEVA ENABLE EMEA 2025. What did you learn about how the two companies are co-innovating products? 

True co-development means deep integration – we essentially become one team, working side by side for three months, six, or even longer to create something truly special. The goal always comes back to: what’s the issue, what’s the value, what are we trying to create for the customer? For example, with the creation of AVEVA’s Industrial AI Assistant, powered by Microsoft, we’re not using technology for the sake of it. We’re answering real customer problems. Collaboratively, we solved problems that were previously unsolvable without this technology. 

AI must be applied to solve specific, contextual problems – not adopted for technology’s sake. We work closely with customers to understand how we can best serve them. Often, we bring the ideas and modules, but our main focus is always on solving a specific business challenge. Furthermore, as technology disruption accelerates, it’s crucial to remember that no one succeeds alone. Building a powerful partner ecosystem is paramount. Disruptive technologies move fast, but collaborative innovation moves much faster. 

AVEVA Enable

Fabio Terezinho speaking at the AVEVA ENABLE EMEA event

What advice would you give to customers who are still wary of sending their on-premises data to the cloud? 

It’s true that some customers still feel their data is safer on-premises. However, I see a consistent and rapid change in customers’ perception, understanding that a software-as-a-service (SaaS) system may offer even more cybersecurity resilience than legacy on-premises systems. Security depends on the physical and cyber protections you have in place – whether on-premises or in the cloud. On-premises systems face their own threats. 

As Marianna Dikaiakou, principal group software engineering manager at Microsoft, told me at this year’s AVEVA ENABLE EMEA: ‘Security is a process, not a product. It’s not something you buy once and forget. It requires continuous education, threat-modelling, patching, and working with responsible partners who prioritise security above all else.’ 

The decision to keep all your infrastructure and data on local servers versus using a SaaS system is like keeping your money under the mattress versus in a bank. And I believe we’ll reach a point where cloud security will be ubiquitous in the same way. Dikaiakou also mentioned the importance of investment and focus on cybersecurity education. I agree – we need to help people understand the measures we take to mitigate risk and that trust can only be built through transparency and consistency. 

How does the partnership between CONNECT and Microsoft showcase the value of interoperability? 

Our CONNECT ecosystem is built on Microsoft Azure, so we inherited scalability without reinventing the wheel. On top of that, we’ve created a specialised layer tailored to industrial data – events, time series, documentation and the associated metadata. That contextualisation is critical: poorly contextualised data produces poor AI results. CONNECT is domain-specific – it's built for industrial use cases with a wide range of tools to gather insights and help users act on critical insights from their data. Microsoft Fabric, meanwhile, is a broader platform AVEVA builds upon. In essence, Microsoft provides the infrastructure and AVEVA brings the industrial expertise. 

The growing demand for interoperability means systems must interact seamlessly. For example, an operational technology (OT) platform improving production efficiency might integrate financial data from SAP. That’s why native connectors are so important. They ensure interoperability without replacing core OT or IT systems, which are each built to solve specific specific business problems.  

How do you see agentic AI evolving, and what role will it play in industry? 

Agentic AI refers to systems made up of AI agents. These systems don’t follow a fixed, deterministic sequence of steps. Instead, they use models, often large language models, to reason about what tools they have, what the user wants, and how best to achieve the goal. They adapt and improve over time, without needing to be reprogrammed or redeployed. Static systems need constant updates from developers, whereas agentic systems learn and evolve on their own. With the right guardrails, these systems can be incredibly powerful. 

What does the next five years have in store for AI and robotics? 

The future includes personal AI agents and intelligent robotics that blend digital and physical capabilities. During our talk, Dikaiakou and I predicted the imminent arrival of Jetsons-style robotic butlers – with any luck, they’ll do our laundry! I think we’ll each have a little group of AI agents helping us navigate our day, making things easier and more efficient and personalised. 

Combining the abstract portion of AI with the physical world of automation and robotics could lead to more autonomous systems, such as those found in industrial automation. Simulations can accelerate research and development, and training models can help robots learn tasks in new environments. As hard as it is to forecast the future landscape accurately, one thing I can predict with certainty is that AVEVA and Microsoft will lead, guide, assist and support our industrial partners and customers through every single step of this journey.  

Watch the interview with Fabio Terezinho and Marianna Dikaiakou to discover how AVEVA and Microsoft are shaping the future of industrial automation with AI-driven innovation 

Discover insights from these partners and more in the Autumn 2025 issue of Technology Record. Don’t miss out – subscribe for free today and get future issues delivered straight to your inbox.     

Contact author

x

Subscribe to the Technology Record newsletter


  • ©2025 Tudor Rose. All Rights Reserved. Technology Record is published by Tudor Rose with the support and guidance of Microsoft.