As an operationally focused area, AI has helped to improve the link between planning and execution. In electrical contracting, AI tools are currently providing more accurate cost and time estimates for completing projects by using past job data, material prices, and labor productivity to produce more realistic projections of what needs to happen. AI is also helping identify potential operational risks, such as supply chain disruptions or crew availability issues, before they become problems at the construction site. Project managers still need to take ownership of their projects' accountability, even though AI can provide suggestions for potential future scenarios. Ultimately, project managers need to verify assumptions, manage relationships with contractors, and make decisions regarding field conditions that may not align with the modeled expectations. Accountability for ensuring the safe completion of projects, creating schedules that work for all stakeholders involved, and communicating with clients cannot be automated.
I believe AI will begin to change the way planning and estimating occur in many ways. Specifically, I believe AI will bring speed and pattern recognition to areas of business that have traditionally relied on gut feeling. Tools can now forecast demand, optimize inventory, and identify potential risks from equipment failures or delivery timing. While AI will provide PMs with clearer, faster decision-making, it will not remove their responsibility to prioritize competing objectives such as cost, quality, and schedule. AI provides information to support decision-making; however, it does not assume responsibility for the decision-making process.
In my opinion, AI is quickly becoming a very effective co-pilot for project planning by simulating numerous outcomes, including cost overruns and schedule delays, and ranking their likelihood. The most significant advantage of this type of technology is its ability to simulate multiple variables in complex programs that would be difficult for a team of humans to comprehend. However, while AI can assist in identifying potential conflicts and ranking their importance, project managers ultimately must assume responsibility for understanding organizational politics, stakeholder sensitivities, and strategic nuances that AI systems do not possess. Additionally, translating AI-identified conflicts into decisions that individuals within the organization will support and implement remains a critical function for human project managers.
From a fintech perspective, AI is shaping the way estimation and risk management are conducted, moving away from a purely reactive, more data-driven approach. As models can stress-test various plans, forecast their potential cash flow implications, and identify hidden risks earlier in the development cycle, PMs will still need to use their judgment under uncertainty. AI is only as good as the data it receives, and it cannot react to sudden changes in regulation, market shocks, or ethical concerns. Therefore, PMs will need to retain governance and oversight responsibilities and make the final decision when the data points under review contradict each other.
With Agentic workflows and generative AI applications at PMs' fingertips, they have access to more tools and information to support better decision-making, increased productivity, and superior traceability. But the fundamentals of ownership and accountability have not changed. While AI tools can help draft a plan, that doesn't mean they can be blindly trusted. You still need to assess the comprehensiveness of real-world assumptions, critical success factors, and dependencies. The estimations still require human judgment, as AI may not be aware of the full context specific to your domain and industry. Risk management is even trickier, as PMs need to ensure that the assumptions captured by AI are complete, without ignoring other new risks arising from AI-controlled workflows, code, and guardrails. PMs must have a thorough understanding of these risks and how to capture them in earlier phases.
With the advent of AI, project managers are no longer required to perform heavy lifting; rather, AI does the majority of that work for them. With the help of AI enabled tools such as Jira AI, ClickUp AI, and Azure DevOps, organizations can now take advantage of the vast amounts of historical sprint data to generate draft roadmaps, effort estimates, and early indication of possible risks. By using these systems, organizations can identify potential areas of schedule slippages, scope creeps or delivery bottlenecks sooner than ever before in the project lifecycle. As a result, the time spent on planning and estimating has been reduced by an estimated 20 to 30 percent while also increasing the accuracy of the forecasts. The project manager still retains ownership of their ability to exercise sound judgement and be accountable for the decisions they make. While artificial intelligence can indicate timelines and identify risks, it cannot tell an organization what constitutes success, assess trade-offs between competing business priorities, or manage stakeholder expectations effectively; this is the responsibility of the project manager. The project manager must make the decision regarding when a change in scope is warranted or necessary due to changing customer value or team capacity. For this reason, successful teams view artificial intelligence as a tool to support their decision making process rather than as an authority to make their decisions for them; this is what enables teams to deliver tangible business value instead of just maximized efficiency in terms of the outputs produced by the project.
AI is currently used to conduct the initial stage of scheduling, forecasting and creating cost estimates, risk breakdowns, etc. For example, a project manager can create a project plan in a matter of minutes instead of several days when using AI tools like this one. AI estimation tools also allow for early detection of potential schedule slipping risks through the application of statistical forecasting methods to similar past projects. While these capabilities improve efficiency for project managers (PMs), they do not alter how PMs determine the prioritization of risks, schedule deviations that may impact a client's trust or the point in the project where the PM must respond to the project scope request from the customer. Ultimately, the PM maintains accountability for prioritizing project risks; placing stakeholder calls; and providing the final decision on whether a change request is accepted or rejected.
I foresee the possibility for AI to make fantastic first drafts on project work by generating plans, estimates, risk lists and option analysis. If you give it prior project data, it could generate a schedule or cost in minutes, therefore saving time. Project managers still however remain accountable for their judgement. While AI provides the project manager insight that a specific task normally slips (along the lines of 10-15%), it cannot stipulate the vendor's reliability (ie Are they reliable? If there are multiple vendors who are equally as viable, which vendor should the project manager select to mitigate vendor risk?) or that the sponsor may panic with just a week's notice of schedule. Good Project Managers have embraced AI as a tool to generate alternative scenarios, allowing them to have a better understanding of potential alternate paths. Once these alternate scenarios are generated, they can use the data available to test their assumptions via modelling based upon people, politics and their experiences. Then they can rely upon their judgement and the data produced by the model to determine which alternative will be taken and communicate the decision to various stakeholders. The overall advantage of this process is enabling the Project Manager to remain accountable for identifying trade-off limitations and communicating those trade-offs; therefore, resulting in fewer project failures and faster project delivery by implementing alternative scenarios.
Artificial intelligence tools are becoming skilled at managing project work in a mechanical capacity. They help Plan, Develop Estimates, Identify Risks by using past patterns, and even recommend Next Steps. This is a time-saving benefit and reduces the number of unrecognized risks, but it is also an area where Project Managers can "lose control" or get off track without careful consideration. AI can help identify project roadblocks; however, AI does not understand the context of these roadblocks; it does not truly understand the trade-offs that must be made; and it will not identify when a team is becoming disengaged from the project or when an apparently "low-risk" project task is politically sensitive. This responsibility remains with the Project Manager. A Project Manager should still take overall responsibility for defining the problem; determining what is important; and making a decision when the information available is incomplete. In addition, they must have the trust of the project team; team members will follow the lead of people, not tools. AI can assist with the decision-making process; however, the final responsibility for those decisions is still retained by the Project Manager, which is a critical distinction.
It's great at pulling data and spotting patterns, and there's no taking that away from AI. The issue is that its forecasts can feel more confident than they actually are. They're based on patterns and past data, not on what's unique about this project, this team, or this moment. And because it lacks that context and even the ability to acknowledge that it's lacking something, it can give you results that aren't the most informed. So if you don't slow down and question the LLM and ask it to find the gaps in its own thinking, then you're obviously not going to have a very accurate result. That critical thinking oversight is what PMs need to own. You still have to decide what actually matters, what can be flexible, and where the real risk is hiding because that insight can't come from a model. In the same vein, you need to own the judgment calls and the conversations that follow, especially when things don't go according to plan.
AI tools are changing planning and estimation processes and risk management and decision-making procedures because they allow teams to analyze data faster and generate precise forecasts and create real-time scenario models which used to require multiple days or weeks of work. Project managers use AI to predict project timelines and detect risks at an early stage which enables them to make better data-based decisions while decreasing uncertainty in complex operational environments that exist in transportation networks which require precise timing and maximum safety and peak cost efficiency. Project managers need to maintain control over the human aspects of leadership which include establishing strategic direction and handling stakeholder relationships and making decisions through their own judgment and conducting ethical accountable decision-making processes when they deal with incomplete or biased AI results.
Artificial intelligence tools have developed new methods for implementing planning, estimation, risk management, and decision-making processes. The planning process benefits from AI, which uses historical project data to determine appropriate project timelines, required resources, and possible project constraints. Machine learning models enable estimation by generating data-driven cost, timeline, and resource requirements predictions that exceed human capabilities. The risk management process uses AI to monitor for irregularities while it forecasts upcoming system failures and identifies hidden system dependencies, which helps teams develop effective risk mitigation strategies. AI enhances decision-making because it can process large volumes of internal and external data to perform scenario analysis, which assists PMs in evaluating various trade-offs and potential results.
We forget about the mountain of invisible admin work that sits quietly on our desks everyday but it's still work that project managers need to get done. Even if it's something as small as tracking the status of projects or flagging surface-level risks, it's hard work and time. And when you can offload that manual effort to AI, you get some breathing room back. And you get back time that can be better used for strategy. So there's definitely value there. But I don't think AI should control the narrative and story of that data, and what it tells you. It can tell you what changed, but it doesn't know why it matters or how to frame it for different people on your team. Deciding what to escalate, what can wait, and how to explain trade-offs to stakeholders is still very much a human task.
AI is changing planning processes. Tools analyze scenarios and identify and evaluate risks buried within the data. They also run estimation models much faster than teams could do manually. This is all very valuable. However, PMs commit a critical error when using AI: They treat AI outputs as decisions rather than as inputs. AI can identify pending launch delay factors and resource bottlenecks from 50 different variables. However, it can't tell you whether on-time shipping is more important than a quality bar, which you need to decide. Your customers need to define it. The relevant PMs leverage AI to reduce uncertainty and guesswork in estimation and risk analysis, then shift their focus to the critical questions that matter: What are we trying to accomplish? Who is this dependent on? What tradeoffs are we willing to accept?
AI has changed how teams plan, estimate, identify and address risks by making the process of finding and understanding that information faster and easier. With AI, teams no longer have to spend hours in meetings or digging through reports to find relevant information. Instead, AI identifies key trends, highlights strategic alternatives and brings to light previously unaddressed areas of concern. It also provides alternative solutions that teams can consider and provides a much clearer picture of what is feasible in their projects. Simulations are the area where AI shows the most improvement. Once all available options have been identified, AI can simulate the impact of each option on the timeline, resources, costs and any early indicators of risk that may arise from implementing that option. As such, teams can make informed decisions using data as opposed to relying on opinion. Nonetheless, while AI can present several options to project managers (PM), ultimately, the decision regarding which direction to take is always left to the discretion of the PM.
We were able to bring down project planning time by nearly 25% once we started using AI tools for budgets and first-pass schedules. However, AI does not run everything in the project. We use these tools to sanity-check estimates and model 'what-if' scenarios, like what would happen if a key location becomes unavailable suddenly. These numbers are usually only as good as the assumptions. On one project, the AI tools gave us a schedule that looked perfect on paper. But, it completely missed the fact there were two scenes that couldn't be shot back to back because of talent shortage, something the PM was able to catch in a matter of minutes. AI is wonderful when it comes to speed and options, but not judgment. PMs will still need to make risk calls, client trade-offs and final sign-offs because if a decision is going to blow the budget by £20k, no tool can make that call, only the PM can and should.
AI has made an enormous difference in how I present information. In the past, I would spend hours pulling reports or creating spreadsheets; now I can generate clear summaries, trend lines, and comparisons in seconds The presentation has also improved. AI allows me to create visual models of what could happen in different scenarios. Visualising these different options before deciding on a course of action provides a much quicker decision-making process. Even though AI provides the information needed to make decisions, ultimately, I still determine which information is most relevant and important, how I want to present the information and what I want my team to focus on.
In my experience AI tools perform exceptionally well for planning and estimation work because they can access previous project information to identify hidden relationships between tasks, discover potential problems at an early stage, and create different scenarios that show how project scope changes will affect results. But I still think great PMs own the "why" and the "so what": they need to establish project priorities, create success criteria, balance project time and quality and budget requirements, and present their choices in a manner that maintains stakeholder understanding and team composure. The AI should create roadmaps and develop timelines and identify risk patterns while you need to verify the results through your judgment and provide additional details which the model cannot understand about political matters and personal relationships and client-specific information before making your decision. The new system allows you to reduce your time on spreadsheets and status reports while you focus on developing conversations and managing expectations and guiding the organization.
Historical pattern analysis by AI tools accelerates planning timelines and workload forecasts. Planning estimation enhance iteration after iteration as the end-of-sprints scope constantly trained the model to "see" resource drift and delivery variance quicker than in an Excel sheet. Risk provision, meanwhile, elevates early warning detection for scope creep, dependency issues, and budget stress indicators. This does indeed give a user some more sort of agent power through a scenario that marks trade-offs the team should reveal but they are too weighted by it. Judgment should be something there to own by PMs because models are based more on the past than the present and its absence with them. Leadership should feel itself needed for alignment of stakeholders, conflict resolution, and ethical accountability. AI cannot sense shifts in team morale due to various reasons including that client politics are similarly in constant motion and making delivery a deadlock, something that AI cannot be finest at understanding. In review, it is up to PMs to define the purpose of the task, validate results as the real target, and not allow the false certainty to slip in. Ownership is determined more by context, priorities, and consequences of an actionable list rather than underlying modeling mechanisms. Artificial intelligence is no more than a calculator in the skilled hands of a strong project manager. In effect, it had appeared that judgment which is continuously formed between intentional context and ether context inevitably grew to reflect proper impairments like closeness and judgment derived from appropriate guidance.
As a marketplace and creative platform, AI is currently transforming the way we approach planning and decision-making by enabling better forecasting and risk assessment in areas that were previously extremely subjective. AI tools can analyze consumer behavior, pricing trends, and other forms of engagement to help teams develop more accurate estimates of future demand, create better-timed, phased product introductions, and identify potential operational risks and mitigate them sooner. Although project managers must continue to maintain ownership of curation, judgment, and strategic direction, AI can identify patterns that cannot determine artistic worth, cultural relevance, or brand integrity.