At the first CTO Connect by Codemotion in Milan, CTOs, Heads of IT, and tech leaders from ENI, Nexi, PagoPA, Fastweb, L’Oréal, Comau, and other leading organizations came together. Four panels focused on AI, security, team building, and digital transformation. Here’s what emerged from the working groups.
When the decision-makers from companies like ENI, Nexi, PagoPA, and Fastweb are in the room, the conversation takes a different turn compared to the usual talks.
No pitches, no demos. Just real discussions about real problems and decisions that affect thousands of developers.
We gathered the most interesting insights from the four themed panels.
AI Agents: Between Experimentation and Production (Spoiler: We’re Still at the Beginning)
Panelists: Massimo Chiariatti (Lenovo), Alessandra Fidanzi (ENI), Stefano Boscolo (CIO Comau), Antonio Iavarone (Iren), Stefano Bussolino (Reply)
Opening question: How many companies are truly putting AI agents into production?
Honest answer: Very few.
Everyone is experimenting—no one can afford to sit still—but actual production use cases are still limited.
The Real Problem
It’s not about the availability of technology. The real challenge is the lack of skills needed to build agents that actually work, integrate them into existing processes, and—most importantly—govern them.
Creating an agent that does something cool in a demo is one thing. Putting it into production, managing its failures, ensuring its security, and making it scalable is something else entirely.
How Agents Are Perceived
An interesting point: Agents aren’t seen as “machines replacing humans,” but as tools that amplify human abilities. The classic augmentation vs. replacement debate. But without structured change management and clear governance, even the most sophisticated agent remains an expensive experiment that brings no value.
Human in the loop was a recurring topic: Humans remain at the center for responsibility, review, and oversight. AI doesn’t decide—it proposes, and humans validate.
The Uncomfortable Question
The most compelling moment of the panel: a provocative question —
When should you NOT use AI?
It may seem obvious, but few actually ask this. Everyone’s rushing to adopt AI, but knowing where it doesn’t make sense can be the difference between a strategic investment and a waste of resources.
Developer takeaway:
If you’re working on AI agents, don’t just focus on “what it can do,” but also ask:
- “How do I integrate it into existing processes?”
- “How do I govern it in production?”
Required skills are changing rapidly.
Security and Governance: The Gap Between What We Think We Have and Reality
Panelists: Dario Carmignani (CRIF), Domenico Raguseo (Exprivia), Giampiero Bonfiglio (L’Oréal), Lorenzo Comi (Nexi)
This panel hit a nerve for many companies who believe they have everything under control when it comes to data security and governance in the AI era.
Spoiler: They often don’t.
The European Gap
It became clear: Europe is still behind the US giants when it comes to cloud architecture and governance.
Many companies are stuck using “single instance” solutions that look enterprise-grade—but aren’t really.
The famous “Secure by Design”? For many, it’s still more of a wish than a reality.
The Cultural Problem
The title said it all: “Secure by design, insecure by culture.”
The tech exists—but company culture often doesn’t keep up.
You can have the most secure architecture in the world, but if your devs haven’t internalized security by design principles, the outcome is fragile.
Investment vs. Cost
A recurring theme: too many companies still approach security with a cost-cutting mindset. They try to save money where they should be investing.
The panel’s provocation:
It’s not about how much it costs to protect yourself, but how much it costs not to.
Developer takeaway:
If you work in environments where security is seen as “overhead” rather than a critical requirement, it might be time to ask the tough questions.
In the AI era, data governance isn’t a nice-to-have — it’s a must-have.
Tech Teams and AI: The Questions No One Wants to Ask (But Everyone’s Thinking)
Panelists: Francesco Dominidiato (MoneyFarm), Francesco Persico (PagoPA), Marco Daccò (RDS), Mirco Patroncini (LastMinute), Layla Pavone (City of Milan)
This was the most direct panel, with the uncomfortable questions that you’re probably wondering about too.
Key Questions Raised:
- Do we still need senior devs when AI writes code faster than anyone?
- Who will train the next generation of developers?
- If AI handles all the repetitive work, how do we build foundational skills?
- Who leads the team when AI does most of the execution?
- And most importantly: How does the CTO’s role evolve when tech starts making decisions on its own?
The Answer Isn’t Just Tech
The solution is not only technological. We need:
- Continuous learning, not just quick certifications or “crash courses”
- Structured mentorship – because that’s how experience is truly passed on
- Cross-functional teams blending tech, language, and critical thinking
- Strategies that align business and innovation, not just deliver features
The Evolving Developer Role
Key point: The developer role is changing rapidly. It’s no longer enough to just write good code.
You need to:
- Understand the business context
- Communicate with non-technical stakeholders
- Think critically about when to use (or avoid) AI
Developer takeaway:
If you think upskilling just means learning the latest framework, you’re missing the big picture.
Soft skills (communication, problem solving, critical thinking) are becoming just as important as technical ones.
AI-Native Startups vs Corporates: Ecosystems That Need to Learn to Speak to Each Other
Panelists: Alexio Cassani (FairMind), Massimo Guarino (Fastweb), Andrea Lonza (Lexroom.ai), Marco Falcone (Indigo), Stefano Priolo (Centrico)
This final panel brought together two worlds that often speak different languages: AI-native startups and large enterprises.
Two Speeds, Same Problems
Startups move fast to stay relevant, trying to create value with limited resources and avoid being “just another wrapper” around existing models.
Corporates need to transform without making mistakes—upskilling at scale and innovating while keeping operations running.
But one shared battleground emerged: data protection, cybersecurity, sustainability, governance.
This is where it’s decided who stays in the game—and who doesn’t.
Make or Buy?
A hot topic: When should you build in-house? When should you buy or integrate existing solutions?
There’s no one-size-fits-all answer, but the consensus was:
Let value drive the decision—not ego or the desire to “do it all yourself.”
Open vs Closed Ecosystems
Another heated discussion: open ecosystems vs proprietary/closed solutions.
Startups lean open, corporates lean toward control. But success will go to whoever can create real value for users.
Final Question
The panel closed with a provocation that left the room in silence:
“Who will still be capable of generating real value in three years?”
Whether you’re in a startup or a corporate, the question is the same. And the answer is far from obvious.
Developer takeaway:
Understanding the startup vs corporate dynamic isn’t just trivia. If you’re in a corporate trying to move like a startup (with enterprise processes), or a startup trying to scale, you live this tension every day. Knowing these dynamics helps you navigate them better.
What We Took Away
CTO Connect wasn’t just another event with slides and presentations.
It was a space where tech leaders from ENI, Nexi, PagoPA, Fastweb, L’Oréal, Comau, and others talked about real problems—unfiltered.
Key insights:
- AI Agents: We’re still at the beginning. The real challenge isn’t building demos—it’s production governance.
- Security: The cultural gap is more dangerous than the tech gap.
- Teams: The developer role is evolving—soft skills are as crucial as code.
- Startup vs Corporate: The real issues (security, governance, value) don’t care about company size.
If you’re a developer, this matters to you.
Because the decisions tech leaders make today are the projects you’ll work on tomorrow.
And understanding where the industry is heading helps you make better career choices.
