It does not replace the responsibility of the engineer or project manager.

Artificial intelligence can accelerate quotes, models, and schedules. On-site, it must remain a tool controlled by professionals. Imagine that an infrastructure department needs to quickly review the technical file of a bridge. Plans are piling up, costs are changing, deadlines are pressing. An artificial intelligence (AI) tool promises to summarize documents, compare variants, and spot inconsistencies. The gain of…

The Editorial of the UNIVGA Review Validation, traceability, contracts, and data must be framed from the start.

Define the authorized tasks: research, pre-writing, comparison, or exploratory calculation.

Assign each final deliverable to an engineer, a BIM manager, or an identified project manager.

Professional Certificate in Planning and Management of Industrial Maintenance Operations

3 months Code: 161619 100% Online

(0)

$380 Certified UNIVGA

Discover the program

What is it, concretely?

Dans le BTP, la modélisation des informations du bâtiment (BIM) rassemble dans une même maquette des plans, matériaux, quantités, coûts, délais et parfois données de maintenance. Les nouveaux outils d’IA promettent d’interroger ces informations, de générer des variantes, de préparer des métrés ou de rédiger des notes de chantier. Certains parlent d’agent d’intelligence artificielle, souvent appelé agent IA, pour désigner un outil capable d’enchaîner plusieurs tâches. Le point clé reste simple : si l’outil se trompe sur une quantité, une règle technique ou une clause, les conséquences peuvent être financières, juridiques et physiques. Bonne nouvelle : bien encadrée, l’IA reste une aide, pas un pilote automatique.

Concrete case: what to do and what not to do

Prenons Buro Happold, bureau international d’ingénierie cité dans les retours de NXT BLD 2026. Le cas suivant est un scénario pédagogique construit à partir des enjeux publics rapportés, pas un incident déclaré. Une équipe prépare les livrables d’un ouvrage : calculs, documents techniques, vérification de maquette BIM et estimation des quantités.

Le problème est classique : il faut produire vite, sans perdre la maîtrise. La tentation consiste à laisser l’IA générer des variantes, des notes de calcul, des métrés ou des comptes rendus. Ce qui manque si l’on va trop vite : une validation humaine claire, un suivi des versions, des règles sur les données, un contrôle qualité et une procédure pour suspendre l’outil s’il devient douteux.

What can be done

Définir les tâches autorisées : recherche, pré-rédaction, comparaison ou calcul exploratoire.
Confier chaque livrable final à un ingénieur, un BIM manager ou un chef de projet identifié.
Conserver les versions des fichiers, des consignes données à l’outil et des résultats obtenus.
Tester l’outil sur d’anciens projets dont les métrés et calculs sont déjà validés.
Prévoir l’arrêt immédiat d’un outil qui produit des réponses fausses ou impossibles à expliquer.

What not to do

Présenter un outil automatique comme un décideur fiable sans expliquer ses limites.
Send the client a note, a measurement, or a model without professional verification.
Submit plans, contracts, or sensitive costs to an uncontrolled online service.
Think that AI reduces the legal responsibility of the company.
Confuse production speed with decision quality on the construction site.

Questions to Ask Before Acting

Does AI only help prepare the work, or does it make a technical decision?
Who signs the final result: the engineer, the project manager, the company, or no one clearly?
Are the data used reliable, recent, and authorized for sharing?
Can we explain how the result was obtained, step by step?
Has the result been compared to a classical method or to a previously validated project?
Does the contract specify the use of AI, data ownership, and responsibilities?
If the tool disappears, changes price, or blocks access, can we recover the project data?

UNIVGA Viewpoint

For Gabon and Francophone Africa, the BIM and AI duo must be treated as a governance issue, not just as a software purchase. Public markets require precisely defined needs and validated preliminary studies according to established procedures. This pushes teams to work with structured, verifiable, and auditable data, especially for bridges, roads, public buildings, and networks.

Data sovereignty also matters. A model may contain plans for hospitals, administrative buildings, works of art, costs, suppliers, or security information. Gabon has a framework for personal data protection and a dedicated authority, the Authority for the Protection of Personal Data and Privacy (APDPVP). Training construction professionals also means training them in confidentiality, contracts, and platform choices.

UNIVGA's position is clear: the future civil engineering professional must become an intelligent controller of AI. They must know how to read a plan, understand a measurement, use BIM, communicate with financiers and lawyers, and then verify the outputs of a digital tool. Open standards like Industry Foundation Classes (IFC), a data exchange format for BIM, can help reduce dependence on closed platforms.

Sources

To go further

10 September 2026

The United Kingdom warns about hidden AI, audit risk

The British NCSC warns about undeclared AI in companies. A key issue for auditing, compliance, data, and governance.

Read the article

4 September 2026

Copyright: Washington supports OpenAI, open risk

Washington supports OpenAI against the New York Times on AI training. For African lawyers, liability remains to be organized.

Read the article

2 September 2026

Google brings YouTube closer to messaging: opportunity and risk

On August 27, 2026, Google announced it is testing YouTube ads that open a conversation with a brand. Useful for selling, to...

Read the article

It does not replace the responsibility of the engineer or project manager.

Offshore: Oxy and Valaris awarded, HSE AI must remain human

MOL solar and battery branch, the cost remains to be proven

Sustainable wood: AFi promotes traceability towards proof

Professional Certificate in Planning and Management of Industrial Maintenance Operations

What is it, concretely?

Concrete case: what to do and what not to do

Questions to Ask Before Acting

UNIVGA Viewpoint

Sources

The United Kingdom warns about hidden AI, audit risk

Copyright: Washington supports OpenAI, open risk

Google brings YouTube closer to messaging: opportunity and risk

Cookies and Privacy

Set cookies

Master of Science (MSc) in Public Law

Shopping cart

Offshore: Oxy and Valaris awarded, HSE AI must remain human

MOL solar and battery branch, the cost remains to be proven

Sustainable wood: AFi promotes traceability towards proof

Professional Certificate in Planning and Management of Industrial Maintenance Operations

What is it, concretely?

Concrete case: what to do and what not to do

Questions to Ask Before Acting

UNIVGA Viewpoint

Sources

Related Articles.

The United Kingdom warns about hidden AI, audit risk

Copyright: Washington supports OpenAI, open risk

Google brings YouTube closer to messaging: opportunity and risk

Shopping cart