AI & Technology Without the Hype

There is a lot of noise around AI in construction and project management right now. Vendors promise automated scheduling. Consultants talk about predictive analytics. Conference speakers suggest that AI will transform how we plan and deliver projects.

Some of this is real. Much of it is overstated.

I use AI tools in my work. I experiment with new technology. I believe there is genuine value in adopting these capabilities thoughtfully. But I also believe we need to be honest about what AI can and cannot do — especially in project planning and controls.

This is not a post about the future. It is about the present. What works today. What does not. And how planners should think about technology adoption without getting distracted by hype.

What AI can realistically help with today

AI is useful for specific, well-defined tasks where pattern recognition, data processing, or text analysis adds value. In project planning and controls, this includes several practical applications.

Schedule review and quality checks. AI can scan a schedule and identify common issues: missing logic, open ends, negative float, unusual durations, activities without resources. This is tedious work for a human. An AI tool can do it in seconds and flag areas for review.

I have used this on programmes with thousands of activities. The AI does not replace the planner's judgement, but it accelerates the review process. It catches things that might otherwise be missed in a manual check.

Risk identification and pattern analysis. AI can analyse historical project data and identify patterns — which types of activities tend to slip, which interfaces cause problems, which assumptions prove unreliable. This is useful for informing risk registers and contingency planning.

The caveat is data quality. If your historical data is incomplete or inconsistent, the patterns will be unreliable. AI does not fix bad data. It just processes it faster.

Reporting and narrative generation. AI can help draft progress narratives, summarise schedule changes, and generate status reports. This saves time on documentation and frees planners to focus on analysis rather than writing.

I use this regularly. It is not about replacing the planner's insight — it is about reducing the administrative burden of turning that insight into a document.

Document analysis and information extraction. AI can read specifications, contracts, and correspondence to extract key dates, constraints, and requirements. This is particularly useful when onboarding to a new project or reviewing large volumes of documentation.

What AI cannot replace

There are things AI cannot do — and will not be able to do for the foreseeable future.

Understanding project context. AI does not understand why a decision was made. It does not know that the client changed their mind last week. It does not understand the political dynamics between the contractor and the subcontractor. It does not know that the site manager is sceptical of the programme and needs to be brought into the process.

This is the critical point. Planning is not just about logic and data. It is about people, relationships, and context. AI cannot navigate a difficult stakeholder conversation. It cannot build trust in a progress meeting. It cannot read the room.

Making judgement calls under uncertainty. AI can process data and identify patterns. It cannot make the judgement call about whether to accelerate a package, accept a delay, or push back on an unrealistic target.

These decisions require experience, intuition, and an understanding of consequences that extends beyond the schedule. A planner who has seen similar situations before can weigh options in ways an AI cannot.

Questioning assumptions. AI works with the inputs it is given. It does not ask whether those inputs are correct. It does not challenge the duration estimate that seems too short. It does not question whether the logic reflects how work will actually be done.

Good planning requires scepticism. It requires asking uncomfortable questions. AI is not sceptical. It processes what it is told.

Building collective understanding. A schedule is not just a technical artefact. It is a communication tool. It represents a shared understanding of how the project will be delivered. Building that understanding requires conversation, negotiation, and collaboration.

AI cannot sit in a room with a project team and build consensus. It cannot explain the critical path to a site supervisor in a way that gets buy-in. It cannot facilitate the planning session that aligns everyone on the delivery strategy.

Why process and thinking still matter more

This is where most teams get stuck when adopting technology. They focus on the tool and forget the process.

AI is an enabler. It can make certain tasks faster and more efficient. But it does not replace the need for clear thinking, structured processes, and disciplined planning practices.

I have seen teams adopt sophisticated scheduling tools while maintaining fundamentally broken planning processes. The tools produced beautiful outputs. The projects still failed. The issue was never the technology.

How planners should approach AI adoption

If you are considering AI tools for planning and controls, here is how I would approach it.

Start with real problems. Do not adopt AI because it is new or because everyone is talking about it. Adopt it because you have a specific problem it can solve. What takes too long? What gets missed? What would improve if you had faster analysis or better data processing?

Be honest about data quality. AI is only as good as the data it works with. If your schedules are inconsistent, your actuals are unreliable, or your historical records are incomplete, fix that first. Technology will not compensate for poor data.

Keep the human in the loop. Use AI to support decisions, not make them. Review what it produces. Question its outputs. Do not assume it is correct because it looks confident. AI can be confidently wrong.

Focus on augmentation, not replacement. The goal is not to replace planners with AI. The goal is to free planners from low-value tasks so they can focus on high-value analysis, stakeholder engagement, and decision-making.

Stay grounded. Ignore the hype. Ignore the vendor promises. Focus on what works today, in your context, on your projects. Be willing to experiment, but be honest about results.

Closing thought

There is a temptation to see AI as a shortcut. A way to do less work. A way to automate planning.

This is the wrong framing.

AI is an enabler. It can help you do certain things faster. It can surface information you might have missed. It can reduce the time spent on documentation and routine checks.

But it does not replace the discipline of planning. It does not replace the need to understand your project. It does not replace the conversations, the judgement, and the experience that turn a schedule into a delivery tool.

The planners who will benefit most from AI are not those who use it to do less. They are those who use it to do more of what matters — analysis, collaboration, decision-making — and less of what does not.

Technology is a tool. Use it wisely. But never forget that the quality of your planning depends on the quality of your thinking, not the sophistication of your software.