The world right now feels like it’s running at two speeds simultaneously — and somehow, neither feels quite right.
On one end, people are panic-adopting every AI tool that drops, rewriting job descriptions overnight, throwing budgets at transformation before anyone asks what exactly they’re transforming into. On the other hand, someone opened ChatGPT once, got a mediocre answer, and decided the whole thing was overhyped.
Both are missing something. And I’ve seen it play out — not just in the headlines, but in every product meeting, every sprint, every digital initiative I’ve been part of.
The Three Camps — And They Exist Inside Companies Too
Axios recently described how AI is fragmenting society into three groups: power users running AI agents around the clock, doubters who still see it as glorified autocomplete, and resisters who understand it well enough to consciously reject it.
What’s less discussed is that this same split is happening inside organizations, and the consequences at the corporate level are more damaging.
Some companies are over-compensating. Feeling the pressure, they rush to “become AI companies,” adding AI to pitch decks, building task forces without the foundational product thinking to make any of it stick. The result is expensive noise: pilots that don’t scale, tools nobody uses, budgets burned without changing how work actually gets done.
Others are the institutional doubters waiting for clearer ROI, for someone else to go first. By the time they move, the gap will be structural, not just strategic.
The companies getting it right are treating AI as infrastructure, not a campaign.
Why the Project Era Is Over
Here’s the model that’s been running digital for decades: define a project, assemble a team, ship a deliverable, disband. Repeat.
Harvard Business Review’s recent research makes the argument clearly — permanent product teams are simply better suited to the pace of digital and AI. The New York Times didn’t turn around its digital business by running projects. It happened when they built persistent teams around a continuous product.
I’ve lived on both sides. The biggest delivery failures I’ve witnessed weren’t technical. They were structural. A project ends, knowledge walks out the door, and six months later, someone is reverse-engineering decisions no one documented.
AI makes this worse. AI-integrated products drift, require monitoring, need retraining, and surface new edge cases constantly. You cannot project-manage your way through that. You need continuous ownership, continuous learning, and continuous iteration.
The Hybrid Truth Nobody Says Out Loud
The answer isn’t “go full AI” or “wait and see.” It’s a hybrid approach that most organizations are too impatient or too risk-averse to actually commit to.
Use AI aggressively where it compounds: data pipelines, lead qualification, content processing, anomaly detection, and customer interaction at scale. The productivity gains aren’t marginal — they’re structural.
But keep humans in the loop where judgment and accountability matter. The organizations that survive this transition will be the ones that used AI to amplify human decision-making, not replace it before they were ready.
A Word to MENA Professionals
We are not observers of this shift. We are inside it.
The advantage of building from this region, with the constraints we operate under, the languages we speak, the problems we actually live with, is that we’re not chasing Silicon Valley’s version of AI. We’re solving for a different context, and that context is genuinely underserved.
My advice: don’t pick a camp. Find the intersection of your existing skills and where AI creates real leverage, and build something there. Something real. Something that runs.
Build something ugly. Ship it. Let a real user tell you it’s broken.
That’s not a slow approach. In this pace of change, it’s actually the fastest one.
Hekmat Ashqar is a Digital Product Leader and AI practitioner based in Ramallah, Palestine. He leads digital product development across mobile, web, and data platforms, and has built AI-powered systems for clients across Europe and the MENA region. He is the co-founder of Sarai, an Arabic NLU automation platform.
