In conversation with Matthias Scharf, Founder & CEO of CPC SRL, an expert in AI-Driven Operations & LegalTech
How can critical infrastructures, data centers, and industrial facilities be operated securely, stably, and efficiently in an increasingly complex world? And what role does artificial intelligence play in minimizing risks, preventing failures, and enabling real-time decision-making?
Matthias Scharf, Founder & CEO of CPC SRL, provides answers to these questions in an exclusive interview with Silicon Valley Europe. Scharf is a recognized expert in AI-driven operations and LegalTech and has made it his mission to transform complex technical systems into scalable, secure, and manageable solutions.
With CPC SRL, he combines hardware, software, and AI analytics into integrated security and operations platforms—from telecommunications and energy networks to data centers, industrial, defense, and logistics facilities. His approach: fragmented data, legacy systems, and isolated monitoring solutions are merged into unified real-time operational intelligence. The result is fewer incidents, significantly reduced monitoring effort, and more informed decisions—exactly when it matters.
A key focus of the discussion is AI video analytics and security integration. As an official Irisity integration partner, CPC SRL demonstrates how modern AI-powered video analytics go far beyond traditional surveillance: from intelligent perimeter and intrusion detection to behavioral analysis and forensic video search, all the way to military-inspired SOC dashboards with real-time situational awareness. Cameras are not just sensors but active decision-making aids—including automatic monitoring of system and camera health.
In the interview, Matthias Scharf discusses why AI is becoming the decisive success factor for critical infrastructures, how security and efficiency can be meaningfully combined, and why clarity from complexity is the key to resilient, future-proof systems. A conversation about technology, responsibility, and the question of how intelligent systems can help relieve human decision-makers—without losing control.
Silicon Valley Europe: Mr. Scharf, you talk about transforming complex systems into scalable and secure solutions. What was the personal trigger for you to focus specifically on this intersection of complexity, security, and AI?
Matthias Scharf: What has always fascinated me is when technology reaches its limits – that’s where the most exciting innovations emerge. In critical infrastructures, data centers, or industrial environments, complexity today is so high that purely manual monitoring and classic IT systems are no longer sufficient to detect risks in time.
I’ve seen in many projects how fragmented systems, silos, and outdated solutions waste valuable time when it really matters. This is where the vision for CPC SRL was born: transforming complexity into clarity by connecting hardware, software, and AI so that operators always have a clear, actionable situational overview – before something becomes critical.
Silicon Valley Europe: CPC SRL operates at the intersection of critical infrastructure, industry, and AI. Why is now the right time to fundamentally rethink existing security and operational systems?
Matthias Scharf: We are experiencing a simultaneous explosion of data volumes and an increase in threats – from cyberattacks to physical security risks. Classic monitoring and control systems are not designed for this; they often operate in isolation and reactively.
At the same time, cloud, edge computing, and AI analytics have advanced to the point where we can now deploy them stably, scalably, and securely in production environments. For operators, this is the opportunity to turn grown silo solutions into integrated platforms: less complexity, more automation, more transparency, and measurably better real-time decisions.
Silicon Valley Europe: Many operators of critical infrastructures struggle with fragmented data and legacy systems. Where do you currently see the greatest risks – and the greatest untapped potential?
Matthias Scharf: The biggest risk is that critical information exists somewhere in the system but no one sees it in time. If video surveillance, access control, sensors, IT monitoring and operational data run separately, the overall picture remains diffuse – in an emergency, every second counts.
The untapped potential lies precisely in this fragmentation: when we unify, correlate and evaluate data streams with AI, many individual signals turn into genuine actionable intelligence. Operators can then detect events earlier, better understand causes and respond more effectively – with less effort in the control room and higher operational safety.
Silicon Valley Europe: You talk about “real-time actionable intelligence.” What does that mean in the daily operations of a data center, energy or telecommunications network operator?
Matthias Scharf: Real-time actionable intelligence means that a system not only collects data but automatically prioritizes, interprets and translates it into concrete action recommendations. For a data center operator, this could mean: the system detects an anomaly at the perimeter, simultaneously checks access log data and camera feeds, and generates a clearly prioritized alarm message instead of simply reporting “movement detected.”
As a result, operators don’t see hundreds of alarms but the three that are truly critical – including context, suggested next steps and seamless documentation. This turns monitoring into active risk management that improves security, availability and efficiency alike.
Silicon Valley Europe: A central element of your solutions is AI-based video analytics. What fundamentally distinguishes modern AI-powered video analytics from classical video surveillance?
Matthias Scharf: Traditional video surveillance means: many cameras, many monitors, and people trying not to miss anything. This is neither scalable nor reliable, because no operator can truly monitor dozens of streams attentively for hours.
Modern AI video analytics turns cameras into intelligent sensors. It detects patterns, behaviors, and anomalies – for example, unauthorized persons at a fence, unusual movement patterns, or abandoned objects – and only reports what is relevant. Combined with forensic video search and customizable detection agents, “watching video” becomes a precise, data-driven security tool.
Silicon Valley Europe: As an official integration partner of Irisity IRIS+, you rely on open integration into existing VMS and camera systems. How important do you consider investment protection when introducing AI systems?
Matthias Scharf: Investment protection is a core principle for us. Many operators have invested in cameras, VMS, and control room technology over the years – the answer cannot be to replace everything. That’s why we integrate Irisity IRIS+ and our own components in a way that allows existing infrastructures to continue being used and intelligently enhanced.
Open interfaces and flexible architectures enable customers to modernize gradually: from the first pilot area to an enterprise-wide AI platform. This reduces project risks, protects budgets, and ensures that AI is seen as an enabler – not a disruptor in ongoing operations.
Silicon Valley Europe: AI can now automatically detect break-ins, unusual behavior, or security-relevant patterns. How do such systems change the role of security and control room personnel?
Matthias Scharf: AI does not replace control room personnel – it finally gives them the tools to maintain an overview in the flood of information. Routine tasks such as continuous screen scanning, simple pattern recognition, or manually searching through hours of video material are automated.
This shifts the role of humans toward evaluation, decision-making, and coordination. Operators focus on truly critical situations, work with clear priorities, and can process incidents more quickly and systematically. This relieves teams, reduces errors, and simultaneously enhances the quality of decisions.
Silicon Valley Europe: You repeatedly emphasize the relief of people in favor of better decisions. Where do you draw the line between automation and human control?
Matthias Scharf: Our philosophy is clear: AI should prepare, prioritize, and recommend – the final decision remains with humans. In critical infrastructures and safety-relevant areas, it is crucial that responsibilities remain traceable and no “black box” feeling arises.
That’s why we place great emphasis on transparent workflows, explainable alerts, and clear escalation paths. The system delivers rich contextual information and action suggestions; the operator decides and documents – supported by the platform, but never replaced. This builds trust and acceptance among all stakeholders.
Silicon Valley Europe: Your solutions are also used in highly sensitive areas such as defense and military. What special requirements do these environments place on AI, security, and system architecture?
Matthias Scharf: In such environments, maximum requirements apply to resilience, availability, and security – both technically and organizationally. Systems must function reliably under stress, during network outages, or in hybrid scenarios combining cloud, edge, and on-premise infrastructure.
At the same time, access control, data classification, and auditability are central themes. We develop architectures that integrate military-grade SOC dashboards, highly secure perimeter protection analytics, and strict compliance requirements – without compromising usability and operational speed.
Silicon Valley Europe: Many companies have reservations about AI, especially when it comes to monitoring, data privacy, and compliance. How does CPC SRL address these concerns?
Matthias Scharf: The reservations are justified – and that’s exactly why we prioritize transparency and privacy-by-design. We approach every project with a focus on data protection, regulatory compliance, and governance: Which data is truly necessary, how is it processed, who has access, and how long is it stored?
Technically, we combine anonymization, role-based access models, and clear logging with organizational measures such as training and policies. This way, AI doesn’t become a risk but a controlled, auditable component of modern security and operational processes.
Silicon Valley Europe: What measurable effects do your customers typically see after implementing your AI and smart systems – for example, in terms of downtime, risks, or operating costs?
Matthias Scharf: Our customers report three main effects: fewer security-relevant incidents, significantly reduced monitoring effort, and improved decision-making quality in the control room. AI-supported perimeter and intrusion detection reduces false alarms, while forensic search and intelligent dashboards drastically shorten incident response times.
Additional savings come from optimized processes and higher utilization of existing infrastructure. When a system detects anomalies early—before they lead to disruptions or failures—unplanned downtimes decrease, which directly impacts operating costs and service quality.
Silicon Valley Europe: Looking ahead to the next five years: How do you see the use of AI in critical infrastructures and industrial environments evolving?
Matthias Scharf: We will see a clear shift from isolated AI islands to end-to-end, platform-based solutions. AI is moving closer to the edge—onto edge devices and cameras—and is simultaneously orchestrated centrally to provide a consistent situational picture across locations and systems.
Moreover, personalized models will gain importance: instead of “one-size-fits-all” algorithms, operators will use AI trained on their specific environments, from the factory gate to the control room. This increases accuracy, reduces false alarms, and transforms AI from an experiment into an indispensable part of operational backbone.
Silicon Valley Europe: Finally, what advice would you give to decision-makers who are currently facing the question of whether and how to integrate AI into their security and operational processes?
Matthias Scharf: Start pragmatically, but with a clear vision. Choose a specific, measurable use case—such as perimeter protection, forensic video search, or consolidating alarm sources—and test how AI supports your teams in day-to-day operations.
Pay attention to three key points: an open, integrable architecture, consistent investment protection, and a partner who understands technology, operations, and regulatory requirements equally. If these fundamentals are in place, a successful pilot can be gradually expanded into a scalable platform that makes your critical infrastructures safer, more efficient, and future-proof.