// Leadership · AI · Strategy · June 2026 · 5 min read

Your AI Strategy Has a
Blind Spot
at the Top.

Pilots everywhere, gains nowhere. ICs can accelerate their own work - only senior leaders can redesign the system around it.

Your company's AI strategy has a blind spot at the top.

Walk through most organizations and AI is everywhere. Pilots, experiments, training. Plenty of activity.

Then leaders look at the business and wonder where the gains went.

01 · Faster tasks, same business

ICs can accelerate their own work with AI.

An engineer can write code faster. A PM can write requirements faster. A designer can prototype faster.

But if the surrounding process stays the same, the business barely moves.

Only senior leaders can redesign the processes, bottlenecks, handoffs, and approvals that actually determine time-to-market. That layer is invisible from the IC seat - and untouchable from it too.

The biggest gains don't come from making individual tasks 20% faster.
They come from rethinking how work flows across teams.

02 · Hands-on is not optional

That kind of redesign requires leaders who are deeply AI-literate and relentlessly hands-on. Not to become the best prompters, but to understand firsthand what AI can and cannot do - and then push the team to apply it.

Reading. Experimenting. Watching demos. Using AI in their own work every week.

Delegating AI fluency while staying detached from it is no longer viable.

Without that firsthand understanding, leaders can't reliably identify:

Which workflows should be redesigned - and which are fine left exactly as they are.
Where weeks can be compressed into days - the handoffs and approvals nobody thinks to question.
Which assumptions about work are no longer true - because the constraint they were built around is gone.
03 · Discovery isn't impact

AI adoption usually starts bottom-up. And it should - experimentation is how organizations discover what's possible.

But discovery isn't impact.

The biggest gains come when leaders take what's proven and redesign the system around it.

You can't redesign
what you don't understand.
The companies that win won't have the
most employees using AI.
Their leaders will understand where AI
changes how the business operates.
ACT ACCORDINGLY...
MF
MARIO FILIPAS
Senior Director, Cloud GPU Software · AMD · University of Waterloo

Leading 150 engineers across Canada, Serbia, and China building GPU virtualization software for AMD's Instinct AI accelerators. I think about what's next...with urgency. I run on AI.

All posts