It’s been a packed week in New York—rain, traffic, and all—but also one full of inspiration and confirmation that the shift we’ve been tracking at The Next Cloud is accelerating fast. I had the chance to attend a closed-door Analyst Summit hosted by Global Relay, the security-first data platform that’s quietly becoming one of the most strategically important companies in enterprise AI.

What stood out wasn’t just the technology—it was the clarity of their vision: clean, structured, private-by-design data as the foundation for AI in regulated industries (and beyond). What they’re building isn’t just another SaaS platform. It’s infrastructure for a new era of AI. An era where autonomy, sovereignty, and trust are more important than scale alone.

That theme repeated itself across the week. First, in Salesforce’s double acquisition. Then, in the news from Canada’s leading telcos. What’s emerging is a clear signal: Agentic AI is coming. AI sovereignty matters. And telcos are ready to play offense.

Act I – The Data Foundation: Global Relay and the Rise of Sovereign AI

At the Global Relay Analyst Summit, one thing became clear: “good data” isn’t just about cleanliness anymore. It’s about control. Enterprises are waking up to the risks of letting their data be used—intentionally or not—for external AI training. If you’re working with OpenAI, Meta, or anyone offering foundation models, you need to ask: what happens to my data?

This is where the term AI sovereignty comes in. It’s not just about compliance—it’s about ownership. Your customer data, your employee communications, your workflows—they’re your competitive edge. You can’t afford to let them leak into the global training corpus.

That’s why platforms like Global Relay are building (their own) AI infrastructure where:

  • The AI comes to the data, not the other way around.
  • Models run on owned or sovereign infrastructure, not public hyperscaler platforms.
  • Everything is auditable, compliant, and private-by-design. Their roots for the past 25 years.

And it’s not just Global Relay. This shift is happening everywhere. And fast.

Act II – Salesforce Doubles Down: From Clean Data to Agentic Execution

Just a few days later, Salesforce made two major announcements:

  • A $8 billion acquisition of Informatica, to deepen its control over data cleanup and governance.
  • A quieter, but arguably more strategic, acquisition of a workflow automation company focused on agentic capabilities—systems that can self-initiate, coordinate, and complete tasks across apps.

Why does this matter?

Because Salesforce sees what many still don’t: that data readiness and execution autonomy are two sides of the same coin. Clean data is the precondition. But the real game is what AI can do with it.

This is the frontier of Agentic AI—where systems don’t just assist; they act. Tools like n8n, Make, MCP (multi-step chain processing), and A2A (agent-to-agent) are enabling AI to move from prediction to orchestration to autonomous action. These aren’t copilots—they’re digital colleagues.

Imagine a future where:

  • A deal isn’t just flagged by AI—it’s reshaped, repriced, and re-routed before you even log in.
  • An anomaly in support isn’t escalated—it’s identified, addressed, and closed with full compliance reporting.
  • Comms platforms don’t just route messages—they trigger workflows, negotiate actions, and optimize in real time.

That’s not science fiction anymore. It’s architecture. It’s design. And it’s moving fast.

Act III – Telcos Join the Race: Infrastructure for the Sovereign AI Era

Then came Act Three—a move that ties everything together.

This week, both Telus and Bell Canada announced major investments in AI data centers across Canada. Telus is spending C$70 billion over the next five years, including new sovereign AI centers in British Columbia and Quebec. Bell launched its AI Fabric initiative, with a network of data centers powered by clean hydro energy, optimized for AI inference workloads.

This isn’t just a telco cloud story. It’s a redefinition of what the network is for.

These sovereign AI factories are:

  • Closer to the edge—where latency matters for real-time, agent-driven systems.
  • Tied to national policy—ensuring compliance with emerging AI regulations.
  • Open to ecosystems—for startups, enterprises, and AI models to co-exist and collaborate without handing over the keys to Big Tech.

And this isn’t isolated to Canada. Globally, telcos are making significant strides:

  • Telstra in Australia has launched its “Connected Future 30” strategy, heavily focused on AI and network intelligence.
  • SK Telecom in South Korea is building modular AI data centers in partnership with Elice and Schneider Electric.
  • Verizon and Lumen in the U.S. are quietly expanding their AI and edge cloud footprint.
  • Adani Group in India plans to invest $10 billion into AI-ready infrastructure, targeting 10 GW of capacity.

This is also a moment of opportunity for ecosystem players. Chipmakers, orchestration vendors, CPaaS providers, data platforms, even regulators—everyone has a role to play in this AI-native, sovereignty-first architecture.

And it’s a wake-up call for anyone waiting on the sidelines. The infrastructure is being built. The models are improving weekly. The regulatory framework is tightening. If you’re not building now, you’ll be stuck integrating later.

Putting It All Together: Why Urgency Matters

What we’re seeing—from Global Relay’s clean AI foundation, to Salesforce’s leap into agentic workflows, to telcos worldwide standing up sovereign AI infrastructure—is the convergence of three mega-trends:

  1. Agentic AI: moving from assistance to autonomy.
  2. AI Sovereignty: keeping enterprise data private, governed, and in-region.
  3. Distributed Cloud Infrastructure: built by telcos, not just hyperscalers.

This is exactly the stack that Que Pasa, CPaaSAA, and others have been preparing for. We already work with the telcos. We’re already plugged into the open network movement with GSMA and Open Gateway. And now, the AI layer is ready to be added—securely, contextually, and with sovereignty by design.

If you’re building in this space—whether you’re a telco, an AI startup, a CPaaS platform, or a chipmaker—this is your moment. Not next year. Now.

Personal Note

Sometimes it feels like I’m watching not just the IT stack, but society itself being re-architected—from the networks that connect us, to the intelligence that powers our decisions, to the infrastructures we now depend on for privacy, trust, and speed.

I wrote a blog not long ago where I said, “My whole life makes sense now”. And this week only reinforced that feeling. Everything I’ve worked on—the dots scattered across semi, cloud, telco, data, AI, communications—they’re all connecting now, fast.

Between the rise of agentic AI, the global momentum around AI sovereignty, and the convergence of telcos, platforms, and cloud communications, we’re entering a phase where technology decisions are – more than ever – also societal decisions.

And I’ll be honest—some nights I can’t sleep. Not from stress, but from excitement. So much is happening. So many stories to tell. So many people to bring together to build what’s next.

That’s why we’re here. To connect the dots. To shape the future—with purpose, speed, and a bit of courage.