AI-Driven Digital

 

Yizhar Mizrachi, CTO and AI Leader at Matrix Digital, reflects on a year in which digital has been reinvented through Generative AI, and outlines the trends that will drive the next wave of innovation in 2026.
An interview.

 

Where Technological Vision Meets Practical Execution

Tell us about your role, what your day-to-day looks like, and what you enjoy most about it.

As CTO and AI Leader at Matrix Digital, my role is to live at the intersection of technological vision and the challenges of real-world execution. My day-to-day life is a roller coaster: I might start the morning with a strategy session on Agentic AI for executives of a large enterprise, and then move directly into a code review or an architectural discussion on microservices with our development teams.

The part I enjoy most? The moment when a buzzword turns into real value. When a solution we designed, whether a complex AI infrastructure or a simple digital interface, meets reality and actually changes the way in which people work or receive services. That’s the moment when technology stops being “cool” and starts being meaningful.

Flagship project: A GenAI chatbot for the Municipality of Ramat Gan – delivered in record time with impressive results: around 95% answer accuracy.

 

Which project or initiative was especially meaningful for you this year?

If you had told me a year ago that we would deploy a fully operational GenAI chatbot for a public-sector organization in just two months, I would have been skeptical. This year, with the Municipality of Ramat Gan, we proved it was possible.

This project set a precedent in Israeli local government: moving from design to production in record time, with about 95% accuracy in responses. But beyond the technology, what made this project truly meaningful for me was its triple impact:

• Technology: This was not a simple keyword-based bot, but an engine that understands intent, cross-references information from multiple sources, and activates APIs in real time.
• Service: Residents stopped “searching for documents” and started “talking to the municipality” instead.
• The call center: AI now acts as a first line of defense, analyzing, explaining, and dramatically reducing the load on human agents.

This project proved to me that even large, complex organizations can achieve fast, smart digital transformation.

The Challenge: Organizational Readiness – GenAI in an Organization Isn’t Like ChatGPT at Home

What do you think was the biggest technological challenge in delivering digital services this year?

The biggest challenge this year wasn’t the technology itself, but rather the ability of organizations to adopt it in a way that is responsible, connected and relevant to everyday operations. Many customers came in expecting an “enterprise ChatGPT”, but quickly realized that GenAI in an organization is something completely different. It’s not just about information security or system stability; it’s a multi-layered challenge combining technological, operational, and organizational complexity:

• The gap between technological capability and organizational reality: Fragmented knowledge, legacy core systems, undocumented processes, and inconsistent data.
• Responsibility and reliability: In an organization, a wrong answer isn’t just an “inaccuracy”; it can lead to incorrect decisions, service failures, or operational risk.
• Changing ways of working: Introducing AI isn’t just a technical move; it reshapes roles, interfaces, and even trust between employees, systems, and customers.
• Expectation management: Understanding that GenAI is not magic, but a capability that requires planning, evaluation, and continuous learning.

This gap between excitement over technological potential and the complexity of actual adoption is the core challenge. Our role is to guide organizations along the way from technological enthusiasm to considered adoption of AI that delivers real business value, that is built on trust, and is sustainable over time.

 

The New Standard: Natural-Language Chatbots; Automated Content Rewriting and Summarizing; and Semantic Search

Which GenAI use cases have become baseline customer expectations this year? And what do you  think has not yet been adopted enough, even though its potential value to organizations is high?

Now considered baseline expectations:

• Natural-language chatbots: Customers expect every website and service to support natural simple, immediate conversational interactions.
• Automated content rewriting and summarizing: Extracting insights and sentiment from large volumes of content.
• Semantic search: People no longer search for “words”; they search for intent.

Still under-adopted, despite high potential:

• Agentic workflows: Digital agents that execute end-to-end processes, delivering dramatic time and cost savings are still in the early stage of adoption.
• AI for internal process optimization: Examples include automated QA, code management, load testing, and automatic system documentation. These are areas with high value, but organizations (especially employees) are concerned that AI will replace them.
• Personalized service at scale: Real-time experience adaptation based on history, context, and preferences. The technology is already mature, but it’s still not adopted by most organizations.

 

From Agentic AI to “LEGO” Platforms: Five Trends Shaping Digital in 2026

Which technology trends from the past year will grow stronger in 2026?

• Agentic AI as standard infrastructure: Organizations will move from GenAI that answers questions to agent-based AI that performs actions. This shift will make every service experience, internal or external, more autonomous.
• Composable digital platforms – “LEGO-style” platforms: Connecting and disassembling capabilities, without major re-architecture. This will be the only way for organizations to remain agile.
• The convergence of UX, Data, and AI: Once separate disciplines, in 2026 they will become a single foundation, with user experiences that are driven by data, context-aware, and responsive in real time.