The AI Efficiency Paradox
AI For Growth Instead of Just Headcount Reduction
Hey, fellow Leader 🚀,
I am Artur, and welcome to my weekly newsletter. I am focusing on topics like IT Management, Innovation, and Leadership, with an Entrepreneurial mindset. My goal is to help you navigate the IT corporate landscape. Make better decisions, create awareness, and share real-world stories.
It has been a wild ride since I published the first article on the SoW. Exchanging views with you and hearing feedback on how these articles have been useful is what drives my motivation to write. Leave a comment, subscribe to the SoW, and be part of the community.
If this article resonates with you, or you know someone who might find it useful, just share the link!
I am writing this article in the midst of Cloudflare’s recent announcement of layoffs affecting 1100 people. AI is widely cited as the primary factor influencing this decision, driven by the technology’s rapid adoption and its subsequent impact on productivity.
Paradoxically, AI is the very reason these roles were eliminated.
Today’s actions are not a cost-cutting exercise or an assessment of individuals’ performance; they are about Cloudflare defining how a world-class, high-growth company operates and creates value in the agentic AI era
However, the overall impact of AI on the labor market remains unclear. While we are undoubtedly at the beginning of a major transformation, the IT job market in both the U.S. and Europe remains stable.
In the US is easier to get data, but evidences show a steady job market in IT despite news of multiple layoffs. In Europe, it is harder to get data specific to IT, however, the unemployment rate remains stable in the euro zone.
Hidden Fallacy on Team Budgeting
Cloudflare won’t be an isolated case. Reducing cost structures due to AI, is becoming a primary goal for companies everywhere. This shift is being enabled by key staff leveraging AI tools, which creates the possibility of leaner organizational models.
However, as talent concentration increases, the structure becomes more fragile: the more centralized the expertise, the higher the impact of a single failure.
If one person with AI is doing the scope of two or three people without AI, the impact of one single point of failure just doubled or tripled.
Who Is Leveraging AI, Really?
This is a question I keep asking myself as I read various use-case articles, and while I analyse and learn the different AI uses. Companies are adding AI into every strategy deck they produce, but it isn’t a tool you can simply broadcast from a mountaintop and expect everyone to capitalize on.
AI is a tool, and like any tool, its effectiveness depends entirely on the person wielding it. Two developers working side-by-side, both powered by AI, can have vastly different performance levels.
This is where the trap lies: you cannot simply mandate AI usage and hope that productivity skyrockets.
There are “star” professionals who have made AI their primary lever, and they are performing brilliantly. Unfortunately, the average use case might have a moderate gain on efficiency while using AI tools like Claude Code. Yet, baselines for cost-effectiveness are being drafted based on these high performers.
AI is not powering teams. AI is powering individuals.
This perspective changes everything.
I recently observed a case where one person handled the workload of two or three different roles, spanning Cloud Engineering to Development. This individual is acing it with an AI tool.
Companies Operate Mostly Understaffed
To say companies operate understaffed is a generalization. Different departments within a large organization often face distinct realities. But as a rule, companies lean toward lean staffing to maximize cost-effectiveness.
This creates constant tension around capacity planning. There is rarely enough FTE availability to meet the demands of the pipeline.
Consequently, prioritization becomes a battlefield, and conflicts arise the moment a stakeholder realizes their request is ranked lower than the “team next door”.
When AI emerged, there was a collective hope that capacity would magically transform. That we could produce features faster with the same number of FTEs.
This would have been a game-changer, providing the margins needed to experiment with new approaches or refine implementations.
Instead, AI is being used as a justification to reduce headcount budgets. In practice, this leaves teams exactly where they were before: understaffed, over-capacity, and drowning in conflicts.
“Why Don’t You Just Use AI”?
The only difference now is the recurring argument for capacity challenges: “Why don’t you just use AI to make it faster?” when this is already the case.
The cycle perpetuates, with the added irony that AI, the tool meant to provide breathing room is being leveraged to shrink the teams even further.
Why Small Isn’t Always Easier
Having managed teams ranging from 3 to nearly 40 FTEs, I can tell you firsthand: managing a small team is often far more difficult than managing a large one.
The primary reason is capacity. When you have less to work with, you spend a disproportionate amount of time negotiating what not to do.
If current trends hold, IT teams will be forced to become leaner and leaner. But small teams face a math problem that AI can’t solve: the work doesn’t adapt to human availability.
In a team of four, if one person takes three weeks for a “once-in-a-lifetime” vacation or maternity/paternity leave, the team instantly loses 25% of its capacity for nearly a month.
Beyond raw capacity, there is the issue of intellectual redundancy.
More Than Just Code
Senior Developers and Architects possess a unique, holistic view of how systems are designed. They are typically the ones best equipped to oversee AI implementations.
In leaner teams, fewer people hold the “big picture” or possess the maturity to “hold the fort” when a key team member is missing. Finding a backup capable of stepping into a Senior’s shoes 100% is rare.
The backup would need to understand both the codebase and the historical business context, all while constantly context-switching to meet demands for his/her own work.
Context-switching used to be a developer’s ultimate productivity killer. Now, it’s risking toward becoming the norm.
Larger teams offer the comfort of overlapping skillsets, which ensures that someone is always there to secure the activity.
Ironically, the way some companies are leveraging AI to justify leaner headcounts is intentionally creating single points of failure and reducing overall systemic resilience.
What Companies Are Leveraging AI For?
This is the fundamental question for assessing how a disruptive tool is being utilized across domains.
Currently, the industry tagline for AI is “increasing productivity and accelerating delivery”. To an extent, this holds true, especially in the beginning when companies decentralize AI usage and star engineers fully harness its potential.
However, my skepticism lies at the leadership level: what will managers actually do with this increased capacity?
Unfortunately, I am seeing a trend where AI is primarily used to suppress operational costs. This includes running cloud systems with fewer resources, avoiding new headcount for critical initiatives, and, increasingly, downsizing existing teams.
Companies lack a proper growth and innovation strategy. When dealing with extended capacity, they lack the vision to understand what to do with it, so they default to cost optimization.
This leads to a pivotal question: If we simply reduce headcount, won’t we just end up producing the same results with fewer people?
If that is the case, the company is negating the true advantages of AI from the start.
This is exactly where I believe an opportunity may arise.
If the overall trend is formed to use AI for reducing the cost structure, the companies that, in contrast, can harness AI to capitalize on existing capacity and find new ways to build and grow, might have found a competitive edge.
Using AI for cost reduction alone will result in static company output supported by leaner cost structures. Most companies are simply using AI to perpetuate the grind with fewer people instead of expanding the ambition. The real winners will be those who use it to build more with the resources they have today.
That’s it. If you find this post useful, please share it with your friends or colleagues who might be interested in this topic. If you would like to see a different angle, suggest it in the comments or send me a message.
Cheers,
Artur
The situation in Europe worries me deeply. IT experts are barely being hired anymore, and IT budgets are being slashed dramatically. The thinking seems to be that this makes companies more competitive — and all that's left to do is send out a memo telling the remaining staff to "just use AI" and everything will carry on as before, only cheaper.
That is a serious mistake.
AI is an opportunity to grow — not a tool to maintain the status quo at a lower cost. And it certainly doesn't implement itself.
As for the mass layoffs at major IT companies, I suspect we're seeing a well-known management pattern play out: cut the headcount, and the people left behind will be forced to figure out a new way to get the job done. They'll come around to using AI on their own. And just like that, management considers the transformation complete.