In prep for a client session later this week, I was re-watching Rob Walker’s keynote at #pegaworld25 — “The Autonomous Enterprise Evolution: From AI to ASI and Back Again” — and again, it struck a chord. Not just because of the big-picture ambition, but because it feels like a conversation we, as business leaders and technologists, need to be having right now. If you haven’t yet seen it, there’s a link at the base, but here’s where I’m at…
Vision of the Future: Exciting… and a Bit Unnerving
Dr Walker opened with the kind of sweeping statement that might sound like sci-fi if it weren’t so grounded in the present: we’re on a path from narrow AI to AGI to ASI — and we need to start asking ourselves whether we’re truly prepared for what’s coming. Yet seasoned followers will appreciate that Rob doesn’t sensationalise: This wasn’t “AI apocalypse” material; It was pragmatic and clear-eyed. He asked a powerful question:
“It’s not just about whether we can build super-intelligence; It’s whether we can keep it aligned with human values.”
And that’s the heart of it: In the rush to automate, optimise, and “agent-ify” everything, we can’t lose sight of the fundamental need for control and ethics.
The bottom line: Agentic AI Is Already Here
While much of the industry is chasing future-state AGI or ASI, it’s important to be mindful that agentic AI is already shaping how enterprises operate.
Personally, I’ve been supporting clients – with Pega technology – for over a decade to deliver capabilities for customer engagement that make decisions in real-time,adapt based on context, and learn and iterate without direct human instruction.
These capabilities are live and driving business outcomes. If you think the AI revolution is something you’ll “get to eventually,” you’re already behind.
GenAI Blueprint: The Missing Link?
We saw a demo of the latest capabilities of Pegasystems GenAI Blueprint, which is something we at DCS have been working with for the past year, demonstrating that it truly is something a lot of enterprises can use today. Our team are using it to support our clients to map and modernise legacy workflows, drop in agentic logic with compliance built-in, and launch fully functional prototypes in a matter of hours – saving weeks in discovery and design.
This bridges a crucial gap: between vision and execution. The primary factor holding most orgs back isn’t ambition — it’s the birds nest of systems and technical debt. Since its introduction, Blueprint felt like an obvious and practical way to finally do something about that.
Guardrails: More Than Just a Buzzword
Where the I’m sure Dr Walker’s keynote will earn even more respect is in its emphasis on governance. Frankly, too many presentations about AI these days focus on acceleration and ignore accountability. Not this one, being clear that autonomous AI without oversight is not innovation — it’s a liability. It’s relevant to all sectors, but particularly in industries such finance, healthcare, and government, where explainable decisions, industry-specific guardrails, and built-in ethics are a must.
It’s easy to get distracted by the hype — by the idea of AI as some distant, transformative silver bullet. But this talk will remind all that the real opportunity is right in front of us. We already have the tools to build adaptive, intelligent, and responsible systems — if we’re willing to roll up our sleeves and truly engage.
If you haven’t already, start now. Stay grounded. And never deploy autonomy without responsibility.
So where next?
Are you already experimenting with agentic AI or Blueprint-style orchestration? I’d love to hear what’s working (or not) in your world.
(And see the keynote here: https://www.pega.com/insights/resources/pegaworld-2025-rob-walker-keynote-autonomous-enterprise-evolution-ai-asi-and)
Rob McLeod
