Artificial intelligence where you can’t see it
Many AI tools that are emerging in the AEC world are not so much about generating shapes or images, but about organizing and interpreting large amounts of design information. The problem is known to those who work in a studio:
- technical details scattered across different files;
- multiple versions of the same documents;
- regulations distributed between texts and subsequent updates;
- knowledge accumulated over the years that remains locked in the archives or in the memory of the most expert professionals.
Some platforms are trying to transform this dispersed material into a searchable knowledge base, capable of retrieving information from previous projects and linking it to new requests. In this sense the studio archive changes nature. It is no longer just a repository of documents, but becomes an active resource that can help reconstruct technical decisions, compare solutions and identify useful precedents.
Much of professional knowledge, in fact, is not found in manuals, but in practices accumulated over time – in files, in construction details, in reused models, in solutions adopted in similar situations. AI works, after all, precisely on this matter.
From database to operational agents
Some developments go even further than simple document searching. More and more often we talk about AI agents: systems capable of autonomously carrying out a series of preparatory operations for the designer’s work. For example:
- search for suppliers for a certain material;
- collect technical data sheets and certifications;
- send quote requests;
- organize the responses received;
- return a comparative summary ready for evaluation.
It is not a question of replacing the designer in decisions, but of reducing the time dedicated to repetitive activities that precede each technical choice. In practice, automating that part of work that today involves online searches, shared folders, emails and phone calls.
A delicate issue: project data
However, there is one aspect that deserves attention. To work well, many of these systems need access to large quantities of internal studio documents: drawings, technical reports, specifications, bills of quantities, correspondence with clients and consultants. This is often sensitive material, which may include economic information, proprietary data, or technical solutions developed over time.
Integrating artificial intelligence into study processes therefore also means addressing a question of data management: where it is processed, who can access it, how it is stored.
For this reason some studios are starting to develop internal tools, built directly on their archives and hosted on local infrastructures. In other cases, the most prudent choice is to use paid professional plans from major AI platforms, which ensure that uploaded data is not used to train models.
Less time searching, more time deciding
If these technologies continue to develop, their most significant impact may not be on the shape of buildings. It could concern the way in which studies organize their knowledge.
Much of the design work consists of looking for information already produced: a detail solved in a previous project, an updated standard, a technical sheet, a solution adopted in a similar case.
If this content becomes accessible and searchable in a more intelligent way, the result is not only an increase in productivity. It also means making more visible that network of knowledge that supports every project: past decisions, consolidated practices, technical compromises.
In the next episodes
In the next articles of this column we will try to go into more detail about these developments. In particular, we will talk about AI agents and how some studios are starting to integrate them into their daily activities: from the search for materials to the management of technical documentation.
These are not futuristic scenarios, but experiments already underway. And probably in the coming years it will be precisely in these less visible spaces – between archives, procedures and information flows – that an important part of the transformation of architects’ work will take place.
The weekly column “Architectural Prompting” is edited by experts Luciana Mastrolia, Giovanna Panucci and Andrea Tinazzo
>> If you are interested in these topics, also sign up to the free Linkedin Newsletter AI & Design for Technicians, we’ll talk about it here!
