Top 10 Ways QS Professionals Use AI on Live Construction Projects in 2026
From Zero to Ten - How AI Became Part of My Daily Professional Life
When I first started using AI tools in my professional QS work, I had one goal. Stop starting from zero.
Every letter, every report, every specification summary, every claim narrative they all began with a blank page. And a blank page, when you are managing multiple subcontracts, responding to Engineer's instructions, preparing payment certificates, and reviewing specifications that run to 500 pages per section, is a genuine productivity problem.
Over time, through trial and error on a live large-scale infrastructure project, I developed a set of practical AI applications that became a consistent part of my daily workflow. Not theoretical. Not experimental. Real applications on real projects, producing real professional output that I reviewed, verified, and issued under my professional responsibility.
This article shares the top 10 ways I and my QS team use AI on live construction projects today. Each one is drawn from direct experience. Each one comes with an honest assessment of where it helps and where professional judgment must still take over.
📚 Related Articles You May Find Useful:
The Top 10 AI Applications in Real QS Practice
1. Professional Letter Drafting
This is the single most consistently valuable AI application in my daily QS work. Letters to the Engineer, variation notices, delay notifications, subcontractor back-charge notices, scope clarification responses, payment correspondence the volume of formal correspondence on a large infrastructure project is significant.
Before AI, every letter started from a blank page. Now I write my key points first, the specific instruction reference, the applicable FIDIC clause, the required action, and then give those points to ChatGPT or Claude to structure into a professionally formatted letter.
A letter that previously took 30 minutes now takes 5 to 10 minutes. Across a project with regular correspondence requirements, that saving accumulates into hours of recovered professional time every week.
Real Example: During the Final Account for an electrical duct bank subcontract, I used Claude to draft a without prejudice settlement letter covering the contractual position, the commercial compromise offer, and the three conditions of settlement. The professional content and judgment were mine. AI structured and formatted them into a professionally presented letter in minutes. See the full story in our Final Account article.
2. Construction Specification Review and Summarization
On my current project, the Technical Specification runs from Section TS 100 to Section TS 700. Each section can be 200 to 500 pages. Reviewing every specification section manually before starting a work package is time consuming but professionally essential.
AI specification summarization has become a standard part of how I approach this. I upload only the relevant specification section to ChatGPT or Claude, ask for a structured summary of the method of measurement, basis of payment, included items, and any "full compensation" clauses, with page and clause references for every point.
The AI summary gives me a structured overview quickly. I then verify every flagged provision against the original specification text before making any commercial decision.
Real Example: A subcontractor submitted a Payment Application billing excavation separately for PVC conduit installation. My specification review of Section 145 found that the linear meter rate constituted full compensation for all works including excavation. The separate excavation billing was rejected with documentary evidence. AI summarization helped me locate the relevant payment clause quickly. See the full story in our Specification Review article.
3. Claim Narrative Report Preparation
A professionally prepared construction claim requires a formal narrative report covering contractual entitlement, cause and effect, delay analysis summary, and quantum. This document is time consuming to prepare and must be professionally structured and formally worded.
My approach: draft my own outline of what the narrative needs to cover first, based on the specific project facts and contract provisions. Then use Claude AI to develop that outline into a complete, professionally formatted narrative report. The professional judgment and factual content come from me. AI develops and structures them into a formally presented document.
Real Example: On the manhole clash variation, I used AI to draft the variation notice and narrative report after the Engineering team confirmed our manholes were built exactly to the Employer design coordinates. The contractual assessment, the evidence compilation, and the cost proposal were professional work. AI drafted the formal narrative efficiently. See the full story in our Variation Management article.
4. Email Management With Outlook Copilot
Every morning at 8:00 AM my working day starts with email. On a large infrastructure project, the inbox contains subcontractor correspondence, Engineer notifications, internal team updates, approval requests, RFIs, and meeting invitations.
Outlook Copilot summarizes my inbox, identifies priority emails, and suggests quick reply options based on email content. Instead of reading every email sequentially and drafting responses from scratch, I get a structured overview and suggested replies that I review, adjust where needed, and send.
The morning email review that previously consumed the first hour of my working day now takes a fraction of that time. See the full daily workflow in our Daily Workflow article.
5. Subcontractor Back-Charge Notice Drafting
Back-charge notices follow a relatively standard structure that changes little between letters. The subcontract clause reference, the description of the event, the back-charge amount, and the formal notification of deduction. Only the specific details change.
Providing the relevant subcontract clause and back-charge details to AI produces a professionally structured draft that requires minimal adjustment before issue. The time saving per letter is not enormous individually but consistent across multiple subcontract packages across a project represents a meaningful reduction in administrative time.
Always verify every AI-generated back-charge letter against your specific subcontract conditions before issue. See the full guide in our Subcontractor Management article.
6. BOQ Description Generation
Generating professionally worded Bill of Quantities item descriptions in accordance with the applicable method of measurement is a time consuming but essential QS task, particularly on tender preparation where time pressure is significant.
AI can generate standard BOQ descriptions efficiently once the work type, applicable measurement standard (NRM, CESMM, or project-specific), and scope parameters are provided. The key prompt discipline: always specify the measurement standard explicitly. AI will not apply NRM or CESMM rules automatically unless instructed.
See our detailed guide on How to Use ChatGPT for BOQ Preparation.
7. Tender Document Summarization
On a one-month infrastructure tender, I spent 8 days reading specification documents, drawings, geotechnical reports, and contract conditions. That 8-day document review is where the commercially significant risks were found, including a Plasticity Index specification that would have been catastrophically expensive if missed.
AI document summarization can compress that initial overview significantly, giving a structured summary of key specification sections in minutes. The detailed professional reading that finds the hidden obligations still requires a qualified QS going through every section carefully. But having that AI overview first helps prioritize where to focus detailed reading attention.
See the full tender preparation workflow in our Tender Preparation article.
8. Progress Report Narrative Generation
Weekly and monthly progress reports require formally structured narrative sections covering overall progress summary, section by section updates, issues and risks, and planned activities. These narrative sections are time consuming to write from scratch every reporting cycle.
Providing the raw progress data to Claude with a structured prompt produces a professionally formatted narrative that I review, verify against the actual site data, and adjust where needed before submission. The time saving on report narrative preparation is consistent across every reporting cycle.
See the full progress reporting workflow in our Progress Reporting article.
9. IPC and Payment Certificate Correspondence
Formal transmittal letters, payment notification correspondence, and query responses related to Interim Payment Certificates can be drafted efficiently using AI once the specific contractual context has been provided. The standard structure of IPC correspondence changes little between payment cycles. Only the specific figures, dates, and reference numbers vary.
Always verify every AI-generated payment correspondence against the specific subcontract payment terms before issue. And never upload sensitive commercial IPC data including subcontract values and payment amounts to free public AI platforms. See the full IPC guide in our IPC article.
10. Professional Qualification Preparation
AI tools supported my MCIOB qualification preparation in three practical ways: grammar checking and writing quality improvement for my experience record submissions, technical topic clarification for competency areas I needed to understand more deeply, and APC interview practice question generation to identify weak areas for development.
The most important warning: never copy and paste AI content into your experience record. Assessors ask detailed follow-up questions about every situation you describe. If the content is AI-generated rather than based on real personal experience, you will not be able to answer those questions convincingly. Write your own experience. Use AI to polish it. See the full guide in our APC Preparation article.
The Complete Picture - What AI Does and Does Not Do
| AI Does This Well | AI Cannot Do This |
|---|---|
| Letter drafting and formatting | Assess contractual entitlement |
| Specification section summarization | Verify site reality and construction facts |
| Claim narrative structuring | Perform delay analysis |
| Email summarization and quick replies | Make commercial judgment calls |
| BOQ description generation | Negotiate with Engineers or subcontractors |
| Progress report narrative drafting | Replace site knowledge and experience |
| Document summarization and overview | Take professional responsibility for output |
| APC interview practice questions | Create your professional experience |
The One Rule That Governs All 10 Applications
Across every one of these 10 applications, one professional rule applies without exception:
Every AI output must be verified by a qualified professional before it is used, issued, or relied upon in any professional context.
AI does not know your specific contract. It does not know your project history. It does not know your site conditions. It does not know your commercial strategy. It may reference clauses that do not exist in your contract, misinterpret technical specification language, or produce output that is professionally inappropriate for your specific situation.
The QS professional who understands this and applies rigorous verification to every AI output gets the full benefit of AI productivity without the professional risk. The QS professional who copies and pastes AI output without checking it will eventually make a costly error.
Use AI as your tool. Keep professional judgment as your foundation. That combination is what makes AI genuinely valuable in construction practice.
Conclusion
Ten practical AI applications. All drawn from real daily professional experience on a live infrastructure project. All verified and applied under professional responsibility.
The common thread across all ten is the same principle my Japanese manager taught me at the very start of my career: TIME IS MONEY. AI saves time on the administrative and documentation aspects of QS work. That saved time goes directly into the commercial judgment, contract expertise, and professional relationships that define the real value of a qualified Quantity Surveyor.
If you are a QS professional who has not yet started using AI tools in your daily work, start with number one on this list. Draft your next letter with AI assistance. Review the output carefully. Issue it under your professional judgment.
Then try number two. Then number three. Build your AI toolkit one application at a time, the same way you built your professional career. One step at a time. Consistently. Without stopping.
The QS profession is not being replaced by AI. It is being upgraded by the professionals who use it well.
💬 Found This Useful?
If this article helped you, please consider:
- ⭐ Sharing on LinkedIn, every QS professional needs to read this in 2026
- 💬 Leaving a comment below, which of these 10 applications are you already using? Which will you try first?
- 📖 Exploring our full blog, each application above links to a detailed dedicated article
Which AI application has made the biggest difference to your QS work? Share it in the comments!
About The Author
A practicing Quantity Surveyor with extensive experience in large scale infrastructure projects under FIDIC conditions. Started as an AutoCAD draftsman at a Japanese construction company. Built a career step by step over 19 years.
Qualifications:
- BSc (Hons) Quantity Surveying
- MCIOB, Chartered Institute of Building
- MQSI, Quantity Surveyor Institute Member
This blog shares real professional experience from live construction projects, not theoretical content.
📚 Related Articles You May Find Useful:
Frequently Asked Questions
What is the most useful AI tool for QS professionals?
Based on direct daily professional use, ChatGPT and Claude AI are the most consistently useful tools for QS correspondence, report generation, and specification summarization. Outlook Copilot is highly valuable for email management if you use Microsoft 365. The best tool depends on the specific task. Build familiarity with two or three tools rather than trying every available option.
How long does it take to learn AI tools for QS work?
With consistent daily practice, most QS professionals develop competent AI prompting skills within two to three weeks. The key discipline is specificity. Always provide detailed context, applicable contract type, required output format, and clear restrictions on what AI should not assume. The improvement in output quality from good prompting is immediate and significant.
Is AI safe to use for construction correspondence?
Yes, with the professional discipline of verifying every AI output against your specific contract conditions before issue. Never upload sensitive commercial data including subcontract values and payment amounts to free public AI platforms. Always check clause references against your actual contract. The QS professional remains responsible for every letter issued under their professional name.
Can AI replace a Quantity Surveyor?
No. AI handles the administrative and documentation aspects of QS work efficiently. The commercial judgment, contract expertise, site knowledge, negotiation capability, and professional accountability that define the QS role cannot be automated. AI makes a competent QS professional significantly more productive. It does not substitute for professional competence.
What is the single most important rule for using AI in QS work?
Verify every AI output before using it professionally. AI does not know your specific contract, your project history, your site conditions, or your commercial strategy. It may misinterpret clauses, overlook important facts, or produce output that is inappropriate for your specific situation. Professional verification is not optional. It is what separates responsible AI use from professional risk.
Where should a QS professional start with AI tools?
Start with letter drafting. Take your next formal correspondence task, write your key points in plain notes, give those notes to ChatGPT or Claude with a structured prompt, review the output carefully, and issue it under your professional judgment. This single application, applied consistently, produces the most immediate and measurable productivity improvement for most QS professionals.
Comments
Post a Comment