The Efficiency Trap: Are We Optimising the Wrong Things in the Age of AI?
Artificial intelligence is helping organisations plan faster, automate processes and analyse more information than ever before. But greater efficiency does not automatically lead to better outcomes.
What happens when we use increasingly powerful tools to accelerate the wrong projects, optimise low-value processes or scale decisions based on incorrect assumptions?
This question was at the centre of “The Efficiency Trap: Are We Optimising the Wrong Things in the Age of AI?”, a panel held on the final day of the Global Project Management Forum 2026 in Riyadh.
The discussion moderated by Youmna Naufal brought together Adrian Predoi, Dave Harry, Aina Aliieva and Ewa Palarczyk. Together, they addressed a challenge that had surfaced repeatedly throughout the forum:
We have developed better tools for doing things faster. But we do not always ask whether those things should be done at all.
As AI increases the speed and scale of execution, the quality of our direction, priorities and decisions becomes more important than ever.
When efficiency lacks direction
Efficiency and effectiveness are not the same.
Efficiency asks how quickly, consistently or economically an activity can be completed. Effectiveness asks a more fundamental question:
Are we achieving the right outcome?
A project can be delivered on time, within budget and according to its original scope, yet still fail to create meaningful value. It may solve the wrong problem, respond to an outdated need or consume resources that could have been used more effectively elsewhere.
AI does not eliminate this risk. It can amplify it.
Automation can make an unnecessary process faster. Advanced reporting can provide greater visibility into a low-value initiative. AI-based tools can improve the execution of a plan that should have been reconsidered before it began.
This is the essence of the efficiency trap: improving the execution of an activity without sufficiently examining whether it is the right activity.
Before asking how AI can accelerate a project, organisations should first ask:
- Does this project address a real and relevant problem?
- Is it aligned with current strategic priorities?
- Are its assumptions supported by reliable data?
- What resources and energy will it consume?
- What meaningful value will it create?
- Should the project be delivered at all?
AI does not create strategic clarity. It amplifies the quality of the priorities, assumptions and decisions already present in the organisation.
When the direction is right, AI can accelerate value creation. When the direction is wrong, it can enable organisations to be wrong at scale, at speed and with confidence.
The portfolio problem
The efficiency trap is particularly visible at portfolio level.
Many organisations run too many projects simultaneously. New initiatives are launched in response to emerging technologies, competitive pressure or changing market expectations. Existing projects remain active because stopping them may be perceived as failure, even when their original business case is no longer valid.
AI can make this environment easier to administer. It can accelerate project initiation, improve tracking and automate status reporting. But more efficient administration does not necessarily create a better portfolio.
The most difficult decisions remain human:
- Which projects genuinely support the organisation’s priorities?
- Which initiatives are competing for the same limited resources?
- Which projects no longer create sufficient value?
- Which activities should be postponed or redesigned?
- Which projects should be stopped?
An organisation can improve the efficiency of every individual project and still maintain an ineffective portfolio.
The real discipline is not doing more things better. It is doing fewer things that matter and giving them the resources, attention and decision-making support required to succeed.
Stopping a project that no longer creates value should not automatically be treated as failure. It can be evidence of responsible leadership and mature portfolio management.
The human cost of constant optimisation
The panel also addressed what organisations risk losing when speed, utilisation and throughput become their dominant measures of success.
Constant optimisation can remove the time needed for reflection, critical thinking and meaningful conversation. Yet these activities often prevent the most serious project failures.
A conversation can reveal that the original brief is based on an incorrect assumption. A moment of reflection can expose a risk that is not visible in a reporting system. A team member who feels able to challenge a decision can prevent an organisation from investing further resources in an initiative that no longer makes sense.
These contributions are difficult to measure through a conventional KPI.
The most important project management activity may sometimes be the conversation that prevents the wrong decision from being made.
Creating space for such conversations is not a rejection of efficiency. It is part of responsible management. Organisations need environments in which people can question assumptions, raise concerns and reconsider priorities without being viewed as obstacles to progress.
AI can provide better information, identify patterns and support scenario analysis. It cannot replace professional judgement, psychological safety or accountability.
From process efficiency to outcome efficiency
The discussion also connects closely with sustainability and energy efficiency.
Improving the efficiency of an individual process is useful, but it is not enough if the wider system continues to consume resources without producing the right outcomes.
Process efficiency asks:
How can we complete this activity using fewer resources?
Outcome efficiency asks:
Are we directing our resources towards activities that create meaningful and lasting value?
Both questions are important.
Sustainable optimisation requires organisations to consider not only how resources are consumed, but also why they are being consumed and what outcomes that consumption creates.
Instead of maximising isolated metrics, organisations need to understand the wider relationship between operational performance, energy use, environmental impact, quality, resilience and long-term strategic value.
Four questions for project and portfolio leaders
Avoiding the efficiency trap does not mean rejecting AI, automation or process improvement. It means applying these capabilities with greater strategic discipline.
Before optimising a project or process, leaders should ask four questions:
1. Are we solving the right problem?
The availability of a new technology should not become the reason for launching an initiative. Start by defining the outcome and identifying the real organisational need.
2. Does this initiative still create sufficient value?
Business conditions, priorities and assumptions change. Projects should be reviewed throughout their lifecycle, not only approved at the beginning.
3. What are we not measuring?
Speed, cost and utilisation do not represent the complete value of an initiative. Consider effects on quality, sustainability, resilience, people and long-term capability.
4. What should we stop doing?
Every organisation has limited resources. Responsible prioritisation requires the courage to pause or discontinue activities that no longer support meaningful objectives.
Efficiency with purpose
The panel did not end with a single answer. It ended with better questions.
That may be its most important conclusion.
The purpose of AI should not be limited to helping organisations perform more activities in less time. Used responsibly, it can support better decisions, reveal dependencies and help organisations use their resources more intelligently.
But technology cannot decide what genuinely matters.
The objective should not be maximum efficiency at any cost. It should be efficiency with purpose, using data, technology, energy and human expertise to achieve outcomes that create real and lasting value.
The question is no longer only whether we can optimise a process.
The question is whether we are optimising the right things.
How can AI support efficiency with purpose?
During GPMF 2026, Ewa Palarczyk presented the original Synergist approach to combining AI, sustainability and energy efficiency.
Discover how technology and data can support better decisions, more responsible resource use and measurable operational improvements:
