Systematics of AI tools supporting creative problem-solving and inventive methods in architectural design

Bartosz Adamiczka

doi:10.37190/arc250110

Abstract

The dynamic development of artificial intelligence (AI) tools significantly impacts contemporary design processes, including architectural design. These tools open new perspectives for creators, supporting them at various stages of their work. AI in architecture is becoming a key element supporting designers in problem-solving, yet there still lacks a systematic approach and classification of related tools in terms of creative problem-solving, inventive methods, and intuitive approach and metaphorical thinking, which are important in the creation process.

The article focuses on the possibilities of using commonly available artificial intelligence tools to support inventive methods in the creative problem-solving process. The author’s first goal is to systematize AI tools in terms of their potential applications in the context of inventive methods. The second is to identify ways to use these tools at individual stages of creative problem-solving.

The analysis was based on data obtained through desk research and classification of tools according to popularity and functionality criteria. Within the study, tools were grouped into five application areas, relating them to successive stages of the creative problem-solving process.

Full article view is only available on bigger screens.

2025

1(81)

Bartosz Adamiczka*

Systematics of AI tools supporting creative problem-solving

and inventive methods in architectural design

DOI: 10.37190/arc250110

Published in open access. CC BY NC ND license

Abstract

The dynamic development of AI tools signicantly inuences contemporary design processes, including architectural design. These tools open

new perspectives for creators, supporting them at various stages of their work. AI is becoming a key element in assisting designers in architecture.

However, there is still a lack of a systematic approach and classication of these tools regarding creative problem-solving, inventive methods, and

intuitive and metaphorical thinking.

This article aims to analyse the potential use of available AI tools to support inventive methods in the creative problem-solving process. The rst

objective is to systematize AI tools with potential applications within inventive methods. The second objective is to identify ways of utilization at

dierent stages of the creative process.

The analysis is based on data gathered through desk research and the classication according to criteria of popularity and functionality. Tools were

grouped into ve application areas, relating them to the stages of the creative process.

The research applied literature analysis, observation, deduction, and synthesis methods. Additionally, the research process was based on inventive

methods: nding connections, alignment between theories, example excursions, mind mapping, deduction and intuitive method that served as the

impulse for conceptual actions at each research stage.

The result is an open systematization of AI tools, enabling designers to compare and select tools that best suit their preferences. The classication,

according to application areas, allows for the selection of a tool based on the nature of the inventive methods used and the stage of the creative prob-

lem-solving process. The developed tabular summaries facilitate a straightforward identication of tools appropriate for specic tasks. The research

ndings highlight the need for further verication of the practical application of AI tools and their incorporation into architectural education, which

will enable a fuller utilization of the potential of these technologies in the design process.

Key words: architectural design, AI tools, inventive methods, creative problem-solving process

Introduction

Society faces a technological revolution, with articial

intelligence (AI) being the primary catalyst for this change.

The introduction of new technologies not only opens the

door to innovation but also raises questions about their im-

pact on society and ethics. At the same time, due to the

rapid pace of technological advancement, the constant

challenge remains functioning in an ever-changing envi-

ronment, where these technologies reshape social values

and establish new ethical standards, creating threats such

as discrimination or privacy violations (Ortega-Bolaños

et al. 2024). Therefore, it is crucial to understand how AI

inuences various aspects of our activities and what mea-

sures can be taken to minimize these potential risks. Equal-

ly important is considering user comfort when utilizing AI

tools and understanding their potential concerns regarding

implementing new systems (Nigam et al. 2021).

In the context of the mentioned challenges, AI is also

revolutionizing the architectural design process by enhanc-

ing eciency, promoting sustainable development, and

fostering creativity (Almaz et al. 2024). AI applications in

architecture include generative design algorithms, virtual

reality tools, and machine learning, enabling architects to

explore innovative design solutions and optimize projects

for energy eciency (Almaz et al. 2024; Chaillou 2021).

* ORCID: 0000-0002-9428-407X. Faculty of Architecture, Wro-

cław University of Science and Technology, Poland, e-mail: bartosz.

adamiczka@pwr.edu.pl

102 Bartosz Adamiczka

Integrating AI with Building Information Modeling (BIM)

allows for real-time analysis and optimization, saving time

and resources (Almaz et al. 2024). The role of AI in archi-

tecture has evolved – from early experiments with comput-

er-aided design to contemporary applications in generative

design and construction robotics (Tellios, Koulali and Val-

samidou 2023). Although AI provides unprecedented re-

sources for design inspiration and parametric exploration,

it is crucial to maintain a sustainable approach to ensure

that AI-generated designs are human-centred, ecological

responsible, and culturally sensitive (Hegazy, Saleh 2023).

Integrating AI in architectural education is also recom-

mended, as it can help designers better understand key

concepts related to the design process (Almaz et al. 2024).

Modern technologies, including AI-powered tools, un-

doubtedly open new perspectives in the design process

– including the search for architectural forms. However,

their implementation should be carried out while pre-

serving elements known from previous design practices.

The key assumption is that these tools should not replace

human creativity but rather complement it, supporting ar-

chitects in creative processes and enriching their potential

(Jaruga-Rozdolska 2022).

The creative thinking process is inherently exploratory,

characterized by variability, dynamism, complexity, and un-

predictability – moreover, it requires an active stance from

the subject (Proctor 2002, 12). At the same time, a far more

comfortable state is the attachment to an established and

familiar paradigm of action based on previously developed

solutions and experiences – which often limits the ability

to discover innovative solutions (Proctor 2002, 29–33).

One of the strategies for designing architectural form can

be the creative problem-solving process (Adamiczka 2022),

which is based on the equal participation of two types of

thinking: reproductive (rational) and productive (creative

intuition). Reproductive thinking utilizes knowledge and

experience, focusing on the analysis and identication of

key aspects of the problem (Proctor 2002, 40, 41); pro -

duc tive thinking, on the other hand, is based on ignorance

– spontaneous and disordered abstractions, verbal associa-

tions, often expressed in the form of images or emotions,

which allows for the generation of new ideas (Proctor 2002,

29–33). The tools that enable the change of the existing par-

adigm and the introduction of new solutions are inventive

methods (Kwiatkowska 2007, 278–283). The combination

of these two types of thinking leads to creativity (Fig. 1).

The aforementioned dynamic development of technolo-

gy, including AI, and the tremendous rise in the popularity

of tools based on it are signicantly changing the possi-

bilities in the design process. As a result, incorporating AI

as one of the potential tools in the creative process is not

only an opportunity but an absolute necessity – especially

when considering the key factor that conditions this pro-

cess: […] the selection of a technique in which the design-

er feels condent – one that will not limit them (Adamiczka

2022, 111). Thinking about contemporary and future gen-

erations utilizing increasingly accessible technologies, it is

clear that the integration of AI-based tools with inventive

methods is, thus, a natural step, enabling the exploration of

new ways of solving design problems.

The main issue addressed in this article is the potential

use of widely available AI-based tools as instruments to

support inventive methods in the process of creative prob-

lem-solving. The rst aim of the discussion is to system-

atize available, commonly used AI tools and their potential

applications in the context of inventive methods. The sec-

ond aim is to identify the possibilities of utilizing AI tools

in the various steps of this process.

Considering the lack of a clearly dened set of widely

available AI tools in the literature, realizing the rst goal

and creating a systematized list of tools in the context of

architecture help raise designers’ awareness of the avail-

able opportunities for their use. Due to the dynamic de-

velopment and technological changes, the list should be

treated as open and an incentive for further exploration

by those interested in the topic. The research adopted two

limitations: 1) the compilation does not serve as a guide

for using individual tools or evaluating their eectiveness;

2) the author focused solely on the process of creative

problem-solving, treated as a path within the previously

described architectural design strategy model

. The analy-

sis addresses the methodology of the process and the sup-

porting tools, leaving the issue of the architectural form

itself beyond the scope of the discussion. This perspective

about achieving the second research goal allows for a fo-

cus on the universal aspects of the process, which can be

applied in various design contexts and serve as a stimulus

for the creative search for problem-solving solutions.

State of research

The topic of AI is currently experiencing exceptional

popularity within specialist circles and a wider audience.

This is not only due to the development of technology it-

self but also its growing impact on daily life and various

professional activities

. The rapid pace of development in

The indicated model of the architectural design strategy, based

on the creative problem-solving process, was described in 2022 in the

doctoral dissertation titled: “The Process of Creative Problem Solving

as a Strategy for Designing an Architectural Form in the Context of the

Phenomenon of Perception” (Adamiczka 2022; in Polish).

The scale and pace of the aforementioned development can be

indicated by the total funding raised by the 50 largest companies in the

AI sector since 2020, which reached $52.8 billion by 2024. Particular-

ly notable was 2023, when a 600% increase was recorded compared to

2020 (Sarkar 2024). This reects one of the key aspects of this techno-

logical transformation: the widespread availability of AI-based tools to

a broad audience, which contributed to their unprecedented popularity.

Between September 2022 and August 2023, over 24 billion visits were

recorded on the 50 most popular AI tool websites (Sarkar 2023). A prime

example is ChatGPT, which has gained tremendous recognition since

its debut in November 2022 and has become one of the fastest-growing

AI-based solutions. Within the rst ve days of its launch, the tool had

1 million users, and by January 2024, this number grew to over 100

million active monthly users. After another nine months, by November

2024, ChatGPT recorded around 200 million active weekly users (com-

pared to 100 million active weekly users in November 2023). The tool’s

popularity was also evident in the number of website visits, which peak-

ed at 1.8 billion monthly. The tool’s success was also reected in the

mobile space – the ChatGPT app for iOS and Android was downloaded

over 100 million times, underscoring its widespread usage (Singh 2024;

Brandl, Cai 2024).

Systematics of AI tools supporting creative problem-solving and inventive methods in architectural design 103

this eld and its broad inuence on many areas of life and

professional activities have contributed to the increased

interest in both the practical use of AI and its theoretical

aspects. This phenomenon is reected in the vast number

of publications in recent years, which focus not only on the

technological development of AI-based specialized tools

but also on analysing the ethical and social consequenc-

es of their use, as well as their presence and application

in education. In 2024 alone, Google Scholar’s database

contains approximately 23,900 records related to the term

“AI-based tools in education”

Similarly, in the eld of architecture, there is an increas-

ing inuence of AI tools, which are widely used both in

generating design concepts and in automating design pro-

cesses. The application of AI in design opens new perspec-

tives but also raises questions about the role of the archi-

tect in the future and the boundaries between creativity and

automation (As, Pal and Basu 2018). Deep learning-based

approaches allow for analysing existing designs, identi-

fying their key elements, and creating new compositions

(As, Pal and Basu 2018). AI is applied at various stages

of the design and implementation process, which opens

up opportunities for automating many aspects of an archi-

tect’s work (Mrosla, Koch and Both 2019). Implementing

AI in architecture inuences not only the design processes

and proposed solutions but also redenes the architect’s

role in future professional practice (Said et al. 2023).

As mentioned, the topic of AI-based tools discussed in

the literature typically focuses on specialized tools and their

echnical aspects. In architecture, the development of tech-

nology is also concentrated on attempts to create tools ca-

pable of solving complex design problems. An example is

a machine learning system based on graphs operating in

three-dimensional space, where the authors highlight the

greater complexity both in terms of structure and combina-

torics compared to systems operating with images, text, or

voice (As, Pal and Basu 2018).

In the context of considerations regarding AI as a tool

supporting inventive methods, it is essential to address

aspects of accessibility and inclusivity. The systematized

tools should not only be available to a wide range of users

but also be simple to operate and free from technological

limitations that could hinder the creative thinking process.

The manner of their use should leave room for intuition –

both to counteract the most destructive beliefs that block

creativity (Proctor 2002, 46) and, as previously mentioned,

the dual nature of thinking in the design process – based

on creative intuition (ignorance and imagination) and ra-

tional thinking (knowledge and understanding) (Kwiat-

kowska 2018, 97–99), as well as for potentially achieving

the most excellent eectiveness of actions (Tarczewski

2019, 49). Considering that the source of creative intuition

is metaphorical thinking (Kwiatkowska 2018, 97–99), an

important role of these tools is to stimulate and inspire the

designer by providing new perspectives, stimuli, and in-

spiration. According to the principles of the creative prob-

Among them, the following works are referenced (As, Pal and

Basu 2018; Mrosla, Koch and Both 2019; Said et al. 2023; Hegazy,

Saleh 2023).

lem-solving process, the methods and tools used within

it should support creativity by oering new possibilities

and facilitating the realization of the designer’s intentions

rather than replacing their role. Their main goal is to assist

in various stages of the design process while maintaining

the key element of human intuition and experience in the

decision-making process (Adamiczka 2022).

Earlier, Hegazy and Saleh, while studying the develop-

ment of AI in architecture, categorize the use of AI tools

into two pathways: parametric exploration and machine

hallucination. The rst pathway relies on the aforemen-

tioned complex tools that consider the detailed parame-

ters input by the designer. In contrast, the second pathway

utilizes […] the ability of a machine or computer program

to generate sensory experiences that do not correspond

to any real-world input and that are not based on any

actual input, but rather are generated entirely by the ma-

chine itself (2023, 12). Regarding the second pathway, the

authors further point out signicant limitations associat-

ed with the use of such tools in the concept creation pro-

cess, including limited control and customization options,

lack of consideration for the feasibility of the structural

design, and inconsistency in the results obtained (Hegazy,

Saleh 2023).

At the same time, the description of the second pathway

corresponds to the previously mentioned metaphorical

thinking. However, Hegazy and Saleh, when describing

and evaluating the discussed tools, apply criteria based on

rational (reproductive) thinking – based on parameters and

measurable real-world characteristics. Referring to the pre-

viously mentioned role of intuition in creative thinking, the

discussion must also be supplemented with the signicant

role of perception in the entire process. This includes, for

Fig. 1. The relationship between rational thinking

and creative intuition in the creative thinking process

(elaborated by B. Adamiczka based on Proctor 2002, 29–33, 40, 41;

Kwiatkowska 2007, 278–283)

Il. 1. Relacja myślenia racjonalnego

i twórczej intuicji w procesie twórczego myślenia

(oprac. B. Adamiczka na podstawie: Proctor 2002, 29–33, 40, 41;

Kwiatkowska 2007, 278–283)

104 Bartosz Adamiczka

instance, the distortion and simplication of complex real-

ities by observers (Lynch 2011, 123), the action of induc-

tion and perceptual forces (Arnheim 2013, 21–23), as well

as the cognitive stratication of what surrounds us – stem-

ming from aesthetic philosophy theory – distinguishing the

physical world, the world of representation, and the world

of assumptions about the reality behind these representa-

tions (Gage 2017).

In light of the above, it must be stated that there is a lack

of sources in the literature that systematically and compre-

hensively describe the approach combining metaphorical

and rational thinking in the context of evaluating AI-based

tools. At the same time, such an application of AI tools

was investigated and described in his doctoral dissertation

Tomasz Broma within the test method of operational trans-

formation of archetypal shapes, comparing the results of

using the inventive method in both traditional and AI-as-

sisted ways (2023, 221–224)

. However, other available

studies, including those referenced in this paper, focus on

analysing AI tools’ technical parameters and measurable

characteristics, neglecting the key role of intuition, per-

ception, and the subjective reception of reality by users

– factors that play a signicant role in the creative pro-

cess. This gap highlights the need to supplement existing

research with a perspective that includes intuitive action,

which could expand the application potential of AI tools in

design practice based on inventive methods.

Furthermore, applying the previously discussed lim-

itation of considering AI tools to those that are widely

available and intuitive to use, it should be noted that there

is a lack in the literature of a comprehensive, systematic

compilation of such tools and their relation to the issue of

the creative problem-solving process.

Methods

The presented research was conducted between Sep-

tember 2022

and October 2024. Key activities were car-

ried out in the third and fourth quarters of 2024. The study

aimed to create an organized collection of AI tools and

identify their potential applications in the creative prob-

lem-solving process. Below, the subsequent steps of the

study are described, considering their sequence and main

assumptions.

At the same time, before proceeding with a detailed de-

scription of the method, it is important to emphasize that

the systematization of AI tools is not only an academic goal

but also a practical support for designers and researchers.

The result of the tool’s operation serves as an inspiration, a rep-

resentation of the abstract understanding of the operation, and a means

to break through initial constraints (Broma 2023, 221–224) for further

design actions – rather than providing a ready-made solution to the de-

sign problem.

Although this case is an exception in the literature, it pertains to

a specic application of a selected and described tool (procedure) and

does not address the systematic classication of AI tools.

This is due to the availability and widespread adoption of the

ChatGPT tool, which has signicantly contributed to the popularization

of AI technology and increased access to information about it (Singh

2024; Brandl, Cai 2024).

Stage 1: Selection of AI tools

As mentioned, the categorized tools should not only be

accessible to a wide range of users but also be easy to use

and free from technological constraints that could hinder

the creative thinking process. The primary factor inuenc-

ing the choice of a tool was the assumption that it should

support the designer, allowing creative freedom through

intuitiveness and ease of use.

In light of the above, the AI tools considered were limit-

ed to the most popular and widely accessible ones based on

the criterion of website trac data for the tools. In this re-

gard, data from an analysis conducted by Sujan Sarkar was

used. Using SEMrush, he analysed information about over

3,000 AI tools gathered from various directories. Based

on this, he identied the 50 most visited tools, which to-

gether accounted for over 80% of the AI industry’s trac

between September 2022 and August 2023 (Sarkar 2023).

The identied 50 AI tools were adopted as the foundation

for the subsequent steps (Table 1, column R), focusing

primarily on AI-based solutions and excluding tools that

utilize AI elements as part of specic functions and serve

only as supplements to a broader range of tools.

Stage 2. Characterization of selected AI tools

In the next step, a characterization of the selected tools

was developed using the desk research method, which in-

cluded the analysis of relevant literature, technical docu-

mentation, APIs, and press materials, including industry

articles and technological reviews. The characterization

was prepared in tabular form. Based on the aforemen-

tioned data, the following aspects were included: tool cat-

egory, popularity (total number of visits from 09.2022 to

08.2023), and website address (Table 1, columns C, D, R)

(Sarkar 2023). At the same time, taking into account the

decision-making process in the design phase regarding the

selection of an AI tool, additional aspects directly related

to the use of the tool were added: main applications, scope

of application, input data type (basic), input data type (ad-

ditional), output data type, user control, languages, hard-

ware requirements, integration with other tools, pricing

model and license, privacy management, and training ma-

terials (Table 1, columns E, F, G, H, I, J, K, L, M, N, O, P).

These aspects were then dened and assigned a specic set

of values, allowing for precise classication and analysis

(Table 1, legend I).

Stage 3: Organizing the AI tools set

by application areas

The list of tools was organized by grouping them based

on their purpose, i.e., tool category and main application

(Table 1, columns C, E). Then, ve application areas were

abstracted (Table 1, legend II, column B). The names of the

areas are characterized by the simplicity that is so import-

ant in the creative thinking process while simultaneously

reecting the basic medium used by the tools. Within each

group, the tools were organized according to the previous-

ly mentioned hierarchy of popularity (Table 1, column D)

Systematics of AI tools supporting creative problem-solving and inventive methods in architectural design 105

and characterized, including a brief description, main ap-

plications, and examples of AI tools for each application

area (Table 1, legend II, columns C, D, E, F). Subsequent-

ly, the specic possibilities for each group to serve as tools

supporting the use of inventive methods were indicated

(Table 1, legend II, column G).

Stage 4: Identification of potential uses

After identifying and describing the areas of application

of AI tools, based on their distinctive capabilities (Table 1,

legend II, column G), the potential uses of these tools with-

in the successive steps of the creative problem-solving pro-

cess were compiled.

For this purpose, the author’s model of the creative

prob lem-solving process pathway, used as a strategy for

de signing an architectural form in the context of its percep-

tion (Fig. 2), was employed. This model aims to stimulate

creative thinking, dene problems, and seek creative solu-

tions regarding architectural form. The developed pathway

aligns with the assumptions of creative problem-solving

theory. It guides the designer through all stages of the pro-

cess – beginning with problem identication, continuing

through the entire design process, and concluding with

the nal version of the project and its verication in terms

of achieving the goal (Adamiczka 2022). Thus, for each

step of the pathway (Table 2, columns A, B) and each area

of application (Table 2, column C), the possibilities for uti-

lizing these tools were identied (Table 2, column D).

Research approach

The above process (stages 1–4) was carried out using

an interactive research and analytical process with the help

of a language model based on articial intelligence to as-

sist in data analysis, classication, and description of AI

tools, using the ChatGPT-4 tool to support exploration and

systematization within the process. The virtual transcript,

totalling approximately 7300 lines

, includes preliminary

instructions, a summary of the conversations, as well as

methodological limitations and research uncertainty, which

was collectively estimated at 10–20%. At the same time,

the conversations were conducted in prototype mode, in-

cluding side and incomplete conversations with the lan-

guage model

. Most importantly, the transcript includes

data and assumption verication using scientic methods

such as synthetic, analytical, deductive, inductive, obser-

vation, and literature analysis. Additionally, the consid-

erations during the process were based on the following

methods rooted in inventive methods: analysis method,

relationship-nding method, congruence between theo-

ries method, example excursion method, mind mapping

method, and deduction method – preceded by an intuitive

The corresponding number of 438,000–584,000 characters with

spaces.

Due to the imperfections (such as suspension and loss of data

from conversations) and the continuous development of the tool (the

availability of a new version of the GPT model required starting a new

chat every day).

method, which served as the stimulus for the actions taken

in the conceptual phase of each stage of the considerations.

Results

The result of the conducted research process is a collec-

tion of tables that together form an open tool supporting

the designer in terms of the potential use of AI tools – both

in the context of their identication, initial exploration, and

selection. Ultimately, with the collected data, it is possible

to make an informed approach to selecting the appropriate

tools based on their functionality and application. Three

areas have been identied in which the results of the work

can provide support to the creator:

1. Identication of the preferred AI tool – the created

tabular compilation of the most popular, widely available,

and intuitive AI tools, along with their parameterized char-

acterization and classication by application areas – pro-

vides the ability to consciously choose a tool that meets the

creator’s needs and preferences (Table 1).

2. Identication of the preferred application area of AI

tools based on the type or nature of the employed inventive

method – a tabular compilation of the designated applica-

tion areas, representing groups of AI tools, by indicating

the characteristic possibilities of their use during the appli-

cation of inventive methods (Table 1, legend II) – provides

the ability for the creator to make a selection based on the

nature of the inventive method used.

3. Identication of the preferred application area of AI

tools based on the stage of the creative problem-solving

process – a tabular compilation of the designated appli-

cation areas, representing groups of AI tools, by relating

them to the creative problem-solving process model along

with indicating their application possibilities (Table 2)

– allows the creator to make a selection based on the stage

of this process.

At the same time, the proposed systematics remains

an open tool – using the presented characterizing aspects

( Table 1, legend I), the creator can consider using an AI

tool outside of the list presented in the paper by comparing

and analysing it with those already included.

Conclusions

The study results conrm the possibility of creating

a practical taxonomy of widely available AI-based tools.

The developed classication of tools represents a signi-

cant contribution to design methodology development, en-

abling more conscious and ecient use of available solu-

tions. A key feature of the created taxonomy is its open

nature, allowing for continuous adaptation to the dynamic

development of AI.

By precisely parameterizing the aspects characterizing

these tools, a comprehensive comparison is made possible,

allowing for an assessment of their functionality in relation

to various aspects of the creative process and the creator’s

preferences. The tabular form of the summaries facilitates

quick and straightforward comparisons of available solu-

tions, supporting the decision-making process in selecting

the tool best suited to current design needs. As a result, this

106 Bartosz Adamiczka

A B C D E F G H I J K L M N O P R

No.

Tool Name

Category

Popularity

(Total Visits bn)

Main

Applications

Scope of

Application

Input Data

Type (Basic)

Input Data

Type

(Additional)

Output Data

Type

User Control

Languages

Hardware

Requirements

Integration

with other

tools

Pricing model

and license

Privacy

Management

Training

Materials

Web Address

Text oriented tools

1 ChatGPT AI Chatbot 14,6 Conversations

and Chatbots,

Text Generation

Universal Text Image Text,

Image

High Multilingual Online Flexible API Freemium,

Subscription, Pay-as-

you-go (Educational)

Stored and Used

for AP (with

User Control)

Documentation,

Video Tutorials,

User Community

chat.

openai.com

2 Character

AI Chatbot 3,8 Conversations

and Chatbots

Specialized Text None Text Medium <10 Online No integration Free Stored and

Shared

Documentation,

User Community

character.ai

3 Quillbot AI Writing 1,1 Text Generation Universal Text None Text Medium Multilingual Online Compatibility Freemium,

Subscription

(Educational)

Stored and Used

for AP

Documentation,

Video Tutorials

quillbot.com

4 Google

Bard

AI Chatbot 0,2416 Conversations

and Chatbots,

Text Generation

Universal Text Image Text Medium Multilingual Online Flexible API Free Stored and

Shared (with

User Control)

Documentation,

Video Tutorials

bard.

google.

com

5 NovelAI AI Writing 0,2387 Creative

Applications,

Text Generation

Specialized Text None Text High 1 (English) Online No integration Subscription Stored and Used

for AP

Documentation novelai.net

6 Janitor AI AI Chatbot 0,1924 Conversations

and Chatbots

Specialized Text None Text Low <10 Online Flexible API Free Stored and

Shared

User Community janitorai.

com

7 You.com AI Chatbot 0,1403 Conversations

and Chatbots

Universal Text None Text Medium Multilingual Online Flexible API Free Stored and Used

for AP (with

User Control)

Documentation,

Video Tutorials

you.com

Perplexity

AI Chatbot 0,1343 Conversations

and Chatbots

Specialized Text None Text Low Multilingual Online No integration Free Stored and Used

for AP

Documentation

perplexity.ai

9 Copy.ai AI Writing 0,1093 Text Generation Universal Text None Text Medium Multilingual Online Compatibility Freemium,

Subscription

Stored and Used

for AP

Documentation,

Video Tutorials,

User Community

copy.ai

10 Jasper AI Writing 0,0949 Text Generation Universal Text None Text Medium <50 Online Flexible API,

Compatibility

Subscription Stored and Used

for AP

Documentation,

Online Courses

jasper.ai

11 Writesonic AI Writing 0,0807 Text Generation Universal Text None Text Medium Multilingual Online Flexible API,

Compatibility

Freemium,

Subscription

(Educational)

Stored and Used

for AP

Documentation,

Video Tutorials,

User Community

writesonic.

com

12 CrushOn

AI Chatbot 0,0246 Conversations

and Chatbots

Specialized Text None Text Low 1 (English) Online No integration Free Stored and

Shared

User Community crushon.ai

13 Claude

AI Writing 0,0212 Text Generation Universal Text None Text Medium Multilingual Online Flexible API Free Stored and Used

for AP

Documentation,

Video Tutorials

claude.ai

14 Personal

AI Chatbot 0,0179 Conversations

and Chatbots

Specialized Text None Text Low <10,

Adaptable

Online Flexible API Freemium,

Subscription

Stored and Used

for AP (with

User Control)

Documentation,

User Community

personal.ai

15 ChatDoc AI Writing 0,0086 Text Generation Specialized Text None Text Medium Multilingual Online No integration Freemium,

Subscription

Stored and Used

for AP

Documentation,

Video Tutorials

chatdoc.

com

Table 1. Classification of selected AI tools (elaborated by B. Adamiczka)

Tabela1. Klasyfikacja wybranych narzędzi AI (oprac. B. Adamiczka)

Systematics of AI tools supporting creative problem-solving and inventive methods in architectural design 107

A B C D E F G H I J K L M N O P R

No.

Tool Name

Category

Popularity

(Total Visits bn)

Main

Applications

Scope

of Application

Input Data

Type (Basic)

Input Data

Type

(Additional)

Output Data

Type

User Control

Languages

Hardware

Requirements

Integration

with other

tools

Pricing model

and license

Privacy

Management

Training

Materials

Web Address

Image oriented tools

MidJourney

Image

Generator

0,5004 Image and

Graphic

Generation

Specialized Text,

Image

None Image High 1 (English) Online No integration Subscription,

Pay-as-you-go

(Educational)

Stored and

Used for AP

User

Community,

Interactive

Materials

midjour-

ney.com

17 CivitAI Image

Generator

0,1772 Image and

Graphic

Generation

Specialized Image None Image High 1 (English) Online Compatibility Free Stored and

Shared

User Community

civitai.com

Cutout.pro

Background

Removal

0,1335 Background

Removal and

Image

Modification

Specialized Image,

Video

None Image Medium <50 Online Compatibility Freemium,

Subscription

Stored and

Shared

Documentation,

Video Tutorials

cutout.pro

19 Craiyon Image

Generator

0,1281 Image and

Graphic

Generation

Specialized Text,

Image

None Image Medium 1 (English) Online No integration Free Stored and

Used for AP

Documentation

craiyon.com

20 Leonardo

Image

Generator

0,1016 Image and

Graphic

Generation

Specialized Text,

Image

None Image High <50 Online /

Local+

(limited

offline)

Compatibility Freemium,

Subscription

Stored and

Used for AP

Documentation,

Video Tutorials

leonardo.ai

21 DeepAI Image

Generator

0,0891 Image and

Graphic

Generation

Specialized Text,

Image

None Image Medium 1 (English) Online Flexible API,

Compatibility

Free (Open-source) Stored and

Used for AP

Documentation,

User Community

deepai.org

Playground

Image

Generator

0,0793 Image and

Graphic

Generation

Specialized Text,

Image

None Image High 1 (English) Online No integration Free Stored and

Used for AP

Documentation playground-

ai.com

23 Stable

Diffusion

Web

Image

Generator

0,0851 Image and

Graphic

Generation

Specialized Text,

Image

None Image High 1 (English) Online /

Online+ /

Local+

(lim. offline)

Flexible API,

Compatibility

Freemium,

Pay-as-you-go

(Open-source)

Stored and

Used for AP

Documentation,

User Community

stabledif-

fusionweb.

com

24 Clipdrop Image

Generator

0,0748 Background

Removal and

Image

Modification

Specialized Image None Image Low <50 Online Compatibility Freemium,

Pay-as-you-go

Stored and

Shared

Documentation,

Interactive

Materials

clipdrop.co

25 Lexica.art Image

Generator

0,0629 Image and

Graphic

Generation

Specialized Text,

Image

None Image Medium 1 (English) Online No integration Free Stored and

Used for AP

Documentation lexica.art

26 PixAI Image

Generator

0,0615 Image and

Graphic

Generation

Specialized Text,

Image

None Image High 1 (English) Online No integration Free Stored and

Used for AP

Documentation pixai.art

27 OpenArt Image

Generator

0,0509

Image and Graphic

Generation

Specialized Text,

Image

None Image High 1 (English) Online No integration Free Stored and

Used for AP

Documentation openart.ai

108 Bartosz Adamiczka

A B C D E F G H I J K L M N O P R

No.

Tool Name

Category

Popularity

(Total Visits bn)

Main

Applications

Scope

of Application

Input Data

Type (Basic)

Input Data

Type

(Additional)

Output Data

Type

User Control

Languages

Hardware

Requirements

Integration

with other

tools

Pricing model

and license

Privacy

Management

Training

Materials

Web Address

Image oriented tools

28 ZMO.AI Image

Generator

0,0496 Image and

Graphic

Generation

Specialized Image None Image Medium <50 Online No integration Freemium,

Pay-as-you-go

Stored and

Shared

Documentation,

Interactive

Materials

zmo.ai

29 Remini Image

Editor

0,0494 Background

Removal and

Image

Modification

Specialized Image None Image Low <50 Online No integration Freemium,

Pay-as-you-go

Stored and

Shared

Documentation remini.ai

30 Dezgo Image

Generator

0,0479 Image and

Graphic

Generation

Specialized Text,

Image

None Image High 1 (English) Online No integration Free Stored and Used

for AP

Documentation dezgo.com

31 VanceAI Image

Generator

0,0417 Background

Removal and

Image

Modification

Specialized Image None Image Medium <50 Online Compatibility Freemium,

Subscription

Stored and

Shared

Documentation,

Video Tutorials

vanceai.com

Video oriented tools

32 CapCut Video

Generator

0,2038 Video and

Audio Editing

Specialized Video Sound Video,

Image

High <50 Local

(online)

Compatibility Freemium,

Subscription

Stored and

Shared

Video Tutorials,

Interactive

Materials

capcut.com

33 Runway

Video

Generator

0,073 Video and

Audio Editing

Specialized Video Image Video High 1 (English) Local

(online) /

Local+

(lim.

offline)

Flexible API,

Compatibility

Freemium,

Subscription

Stored and Used

for AP

Documentation,

Video Tutorials,

User Community

runwayml.

com

34 D-ID Video

Generator

0,0673 Video and

Audio Editing,

Creative

Applications

Specialized Video,

Sound

Image Video,

Sound

High <50 Online Flexible API Subscription Stored and

Shared

Documentation,

Video Tutorials

d-id.com

Synthesia

Video

Generator

0,0582 Video and

Audio Editing,

Teaching and

Education

Specialized Video,

Sound

Text Video High Multilingual Online Flexible API Subscription

(Educational)

Stored and Used

for AP

Documentation,

Online Courses,

Interactive

Materials

synthesia.io

36 Kaiber Image &

Video

Generator

0,0476 Video and

Audio Editing,

Image and

Graphic

Generation

Specialized Video,

Image

Text Video,

Image

High <50 Online No integration Freemium,

Subscription

Stored and Used

for AP

Documentation kaiber.ai

37 Opus.pro Video

Generator

0,0224 Video and

Audio Editing

Specialized Video Sound Video High <50 Online No integration Subscription Stored and Used

for AP

Documentation,

Video Tutorials

opus.pro

Table 1 continued. Classification of selected AI tools (elaborated by B. Adamiczka)

Tabela 1 cd. Klasyfikacja wybranych narzędzi AI (oprac. B. Adamiczka)

Systematics of AI tools supporting creative problem-solving and inventive methods in architectural design 109

A B C D E F G H I J K L M N O P R

No.

Tool Name

Category

Popularity

(Total Visits bn)

Main

Applications

Scope

of Application

Input Data

Type (Basic)

Input Data

Type

(Additional)

Output Data

Type

User Control

Languages

Hardware

Requirements

Integration

with other

tools

Pricing model

and license

Privacy

Management

Training

Materials

Web Address

Sound oriented tools

38 Vocal

Remover

Voice &

Music

0,1655 Other

Specialized

(Sound Analysis,

Vocal Removal)

Specialized Sound None Sound Low 1 (English) Online No integration Free Stored and Used

for AP

Documentation vocalre-

mover.org

Eleven Labs

Voice &

Music

0,0886 Other

Specialized

(Voice Synthesis)

Specialized Sound None Sound High <50 Online No integration Freemium,

Pay-as-you-go

Stored and Used

for AP

Documentation eleven-

labs.io

40 Voicemod Voice &

Music

0,0735 Other

Specialized

(Voice Effects)

Specialized Sound None Sound Medium 1 (English) Local

(online)

No integration Freemium,

Subscription

Stored and

Shared

Documentation,

Video Tutorials

voicemod.net

41 Otter.ai Voice &

Music

0,0707 Teaching and

Education

(Transcription

Support)

Universal Sound Text Text Low <50 Online Compatibility Freemium,

Subscription

(Educational)

Stored and Used

for AP (with

User Control)

Documentation,

Video Tutorials

otter.ai

42 Noty.ai Voice &

Music

0,009 Teaching and

Education

(Transcription

Support)

Universal Sound Text Text Low <50 Online Compatibility Freemium,

Subscription

Stored and Used

for AP

Documentation,

Video Tutorials

noty.ai

Automation oriented tools

43 Hugging

Face

Data

Science

0,3166 Data Analysis

and AI Models

Specialized

Text,

Tabular

Data

Complex

Tabular

Data,

Text

High <10,

Adaptable

Online /

Local +

(limited

offline)

Flexible API,

Compatibility

Free,

Pay-as-you-go

(Educational,

Open-source)

Stored and Used

for AP (with

User Control)

Documentation,

Online Courses,

User Community

hugging-

face.co

44 Hotpot.ai Design 0,1253

Image and Graphic

Generation,

Creative

Applications

Specialized Text,

Image

None Image High <50 Online Compatibility Freemium,

Subscription

Stored and Used

for AP

Documentation,

Video Tutorials

hotpot.ai

45 Tome Design 0,0884 Creative

Applications

Universal Text Image Text,

Image

Medium <50 Online No integration Freemium,

Subscription

Stored and Used

for AP

Documentation,

Video Tutorials

tome.app

Eightfold

Recruiter 0,0772 Process

Automation

Specialized

Tabular

Data

Text

Tabular

Data

Low <50 Online Compatibility Subscription Stored and Used

for AP

Documentation,

Video Tutorials

eightfold.ai

47 Photo

-Room

Design 0,0661 Background

Removal and

Image

Modification

Specialized Image None Image Low <50 Online Compatibility Freemium,

Pay-as-you-go

Stored and

Shared

Documentation,

Interactive

Materials

photo-

room.com

48 Zyro Design 0,0601 Creative

Applications

Universal Text Image Text,

Image

Medium <50 Online Compatibility Subscription

(Educational)

Stored and

Shared

Documentation,

Video Tutorials

zyro.com

49 Gamma Design 0,0276 Creative

Applications

Universal Text Image Text,

Image

Medium <50 Online No integration Freemium,

Subscription

Stored and Used

for AP

Documentation,

Interactive

Materials

gamma.app

50 Taskade Task

Manager

0,0231 Process

Automation

Universal Text None Text Medium Multilingual Online Compatibility Freemium,

Subscription

Stored and Used

for AP

Documentation,

Video Tutorials

taskade.com

110 Bartosz Adamiczka

A B C

Name of the Characterizing

Aspect

Definition / Description

of the Characterizing Aspect

Set of Values

Main Applications Defines the key areas where the AI tool

finds practical application. It indicates

the type of activities the tool supports

and the fields in which it can be utilized.

Text generation – creating, editing, paraphrasing, and synthesizing texts.

Conversations and chatbots – interactive dialogues and customer support, personal chatbots.

Image and graphics generation – creating images from descriptions, generative art, graphic design.

Video and audio editing – creating or editing multimedia content.

Process automation – tools supporting business processes such as recruitment, task management.

Data and AI model analysis – platforms supporting the building, testing, and deployment of AI models.

Background removal and image modification – precise removal of image elements or editing details.

Creative applications – generating artistic, literary, or visual content.

Teaching and education – creating educational materials, language learning support, or generating interactive content.

Other specialized – tools with unique functions, such as sound analysis or vocal removal from songs.

Scope

of Application

Defines whether the tool has a wide

application across various fields (universal)

or is dedicated to specific tasks (specialized).

Universal – tools with a wide range of functions, enabling use across various fields and tasks without the need for specialized knowledge.

Specialized – tools designed for specific applications or industries,

focused on particular functions that require knowledge and have limited functions outside the target application.

Input Data Type (Basic) Defines the basic type of data that the user

inputs into the tool to obtain the expected

results.

Text – written data, such as queries, paragraphs, or documents, processed for content generation or analysis.

Image – graphic files, photos, or illustrations used for analysis, editing, or generation.

Video – films or animations subjected to editing, analysis, or new content generation.

Sound – audio recordings, music, or other sounds, transformed or analyzed.

Tabular Data – structured data in tables, such as spreadsheets or CSV files, used for analysis or modeling.

3D Model – spatial three-dimensional data used for rendering, analysis, or animation.

Complex – data combining various formats, such as text, images, video, or sound, for multidimensional processing.

Input Data Type (Additional) Defines the additional type of data that

the user inputs into the tool to obtain

the expected results

(as attachments or additional, less frequently

used data formats).

Text – written data, such as queries, paragraphs, or documents, processed for content generation or analysis.

Image – graphic files, photos, or illustrations used for analysis, editing, or generation.

Video – films or animations subjected to editing, analysis, or new content generation.

Sound – audio recordings, music, or other sounds, transformed or analyzed.

Tabular Data – structured data in tables, such as spreadsheets or CSV files, used for analysis or modeling.

3D Model – spatial three-dimensional data used for rendering, analysis, or animation.

Complex – data combining various formats, such as text, images, video, or sound, for multidimensional processing.

Table 1. Legend I: Characteristic aspects of AI tools (elaborated by B. Adamiczka)

Tabela 1. Legenda I: Aspekty charakteryzujące narzędzia AI (oprac. B. Adamiczka)

Systematics of AI tools supporting creative problem-solving and inventive methods in architectural design 111

A B C

Name of the Characterizing

Aspect

Definition / Description

of the Characterizing Aspect

Set of Values

Output Data Type Defines the type of results generated by the

AI tool in response to the user’s

input data.

Text – written data, such as queries, paragraphs, or documents, processed for content generation or analysis.

Image – graphic files, photos, or illustrations used for analysis, editing, or generation.

Video – films or animations subjected to editing, analysis, or new content generation.

Sound – audio recordings, music, or other sounds, transformed or analyzed.

Tabular Data – structured data in tables, such as spreadsheets or CSV files, used for analysis or modeling.

3D Model – spatial three-dimensional data used for rendering, analysis, or animation.

Complex – data combining various formats, such as text, images, video, or sound, for multidimensional processing.

User Control Defines the approximate degree to which the

user can influence the result through person-

alization, parameter selection,

or iterative refinement of outcomes.

Low – lack of or limited ability to adjust parameters or iteratively refine results.

Medium – moderate control with customization options, such as tone, style, or detailed prompts.

High – significant control over results with detailed parameters, editing tools, or iterative improvements.

Hardware Requirements Defines whether the tool requires

an internet connection, whether

the software needs to be downloaded

and installed on the user’s device,

and whether advanced hardware

is required.

Online – the tool operates entirely in a web browser or through an online application;

requires an internet connection; does not require advanced user hardware.

Online+ – the tool operates entirely in a web browser or through an online application; requires an internet connection; requires advanced user hardware.

Local (online) – the tool requires downloading and installing software on the user’s device; requires an internet connection;

does not require advanced user hardware.

Local+ (online) – the tool requires downloading and installing software on the user’s device;

requires an internet connection; requires advanced user hardware.

Local (limited offline) – the tool requires downloading and installing software on the user’s device; does not require an internet connection;

limited functionality without an internet connection; does not require advanced user hardware.

Local+ (limited offline) – the tool requires downloading and installing software on the user’s device; does not require an internet connection;

limited functionality without an internet connection; requires advanced user hardware.

Local (offline) – the tool requires downloading and installing software on the user’s device; does not require an internet connection;

full functionality without an internet connection; does not require advanced user hardware.

Local+ (offline) – the tool requires downloading and installing software on the user’s device; does not require an internet connection;

full functionality without an internet connection; requires advanced user hardware.

Languages Defines the range of languages supported by

the tool, including multilingual

capabilities and the ability to adapt to

specific languages.

1 – the tool supports only one language.

<10 – the tool supports between 2 and 10 languages, offering limited multilingual support.

<50 – the tool supports between 11 and 49 languages, providing support for different regions and basic multilingual functionality.

Multilingual – the tool supports over 50 languages, ensuring broad global availability.

Adaptable – the tool allows for adding support for new languages through model training or introducing new data.

112 Bartosz Adamiczka

A B C

Name of the Characterizing

Aspect

Definition / Description

of the Characterizing Aspect

Set of Values

Pricing model and license Defines the type of access to the tool,

including its cost

(with reference to educational licenses)

and the nature of the license

(with reference to open-source licenses).

Free – the tool is fully accessible without charges for the end user.

Freemium – basic features of the tool are available for free,

but advanced features require payment.

Paid: One-time payment – requires a one-time license purchase,

granting permanent access to features without further fees.

Paid: Subscription – the user must regularly pay a fee

(e.g., monthly, annually) for access to features.

Paid: Pay-as-you-go – the user pays for each unit of use

(e.g., processed data, generated content, or minutes).

Integration with other tools Defines the ability to collaborate

with other systems and the availability

of an API interface,

enabling integration

with external applications.

No integration – the tool operates in isolation,

with no integration capabilities with other platforms.

Compatibility – the tool works with other popular systems or platforms

(ready-to-use integrations for non-technical users with commonly used platforms).

API flexibility – the tool offers a flexible API interface for easy integration with external applications

(deep, flexible, and customizable integration for developers).

Training Materials Defines the type and scope

of available basic educational materials that

support the user in learning

how to use the tool.

Documentation – official documents describing features,

user manuals, and technical specifications.

Video tutorials – a set of video materials demonstrating

how to use the tool.

Online courses – comprehensive educational courses

conducted on dedicated platforms or within the tool itself.

Webinars – online meetings conducted by experts

to familiarize users with the tool’s features or methods of use.

User community – forums, discussion groups,

or communities supporting the exchange of knowledge and experiences.

Interactive materials – exercises, simulations,

or interactive tutorials available within the tool.

No materials – no available educational materials.

Privacy Management Defines how the tool manages user data,

including storage, sharing,

and privacy protection

(with user control included).

Stored and shared – user data is stored and may be shared with third parties.

Stored and used for AP – data is stored and used for analysis but not shared.

Not stored – data is not stored after the user session ends.

Table 1. Legend I continued: Characteristic aspects of AI tools (elaborated by B. Adamiczka)

Tabela1 Legenda I cd.: Aspekty charakteryzujące narzędzia AI (oprac. B. Adamiczka)

Systematics of AI tools supporting creative problem-solving and inventive methods in architectural design 113

A B C D E F G

No.

Application Area Description Main Applications Examples of AI tools

Number of

tools

Characteristic application possibilities

in the use of inventive methods

Text

oriented tools

Tools supporting natural communication

in the form of conversations and text generation,

ranging from short responses to complex creative

and functional content.

Used in chatbots, voice assistants,

marketing, and writing.

Conversations and Chatbots,

Text Generation,

Creative Applications

ChatGPT, Character AI, Quillbot,

Google Bard, NovelAI, Janitor AI, You.com,

Perplexity AI, Copy.ai, Jasper, Writesonic,

CrushOn AI, Claude AI, Personal AI,

ChatDoc

Generating concept lists,

paraphrasing and transforming text,

creating metaphors and analogies,

analyzing and synthesizing information,

iterative refinement,

supporting the narrative process,

automatic question generation.

Image

oriented tools

Tools for generating, editing,

and modifying images,

including background removal

and graphic enhancement.

Useful in design, digital art,

and visual marketing.

Image and Graphic Generation,

Background Removal

and Image Modification

MidJourney, CivitAI, Cutout.pro, Craiyon,

Leonardo AI, DeepAI, Playground AI,

Stable Diffusion Web, Clipdrop, Lexica.art,

PixAI, OpenArt, ZMO.AI, Remini,

Dezgo, VanceAI

Text to image transformation,

creating variations and transformations

of images,

combining multiple images or fragments,

simulating effects and styles,

removing or editing selected elements.

Automation

oriented tools

Tools that combine creativity

with process automation,

supporting the creation of presentations,

visual materials, and task

and data management.

They enhance the efficiency of both creative

and operational work.

Creative Applications,

Image and Graphic Generation,

Data Analysis and AI Models,

Process Automation,

Other Specialized

(Recruitment, Task Management)

Hugging Face, Hotpot.ai, Tome,

Eightfold AI, PhotoRoom, Zyro,

Gamma, Taskade

Automatic generation of proposal lists,

streamlining data analysis,

automatic prototyping and iteration,

automation of repetitive tasks,

creating graphical representations of data sets,

optimization of evaluation processes.

Video

oriented tools

Tools for video and audio editing,

creating multimedia content,

and supporting creative projects

in education and marketing.

They facilitate the production

of professional audiovisual materials.

Video and Audio Editing,

Creative Applications,

Teaching and Education

CapCut, Runway ML, D-ID,

Synthesia, Kaiber, Opus.pro

Transforming text into animation,

animating still images,

combining and mixing multiple materials

or segments, simulating effects and styles,

automating clip and narration editing.

Sound

oriented tools

Tools specialized in sound processing,

voice synthesis, music analysis,

and speech transcription.

They are used in education, music production,

and sound effect creation.

Other Specialized

(sound analysis, vocal removal,

voice synthesis),

Teaching and Education

(transcription support)

Vocal Remover, ElevenLabs,

Voicemod, Otter.ai, Noty.ai

Generating new sounds and effects,

transforming and modifying sounds,

removing and isolating elements,

creating synthesized voices,

automatic transcription.

Table 1. Legend II: Areas of application of AI tools with indication of characteristic application possibilities in the use of inventive methods (elaborated by B. Adamiczka)

Tabela1 Legenda II: Obszary zastosowania narzędzi AI ze wskazaniem charakterystycznych możliwości ich zastosowania w toku wykorzystania metod inwentycznych (oprac. B. Adamiczka)

114 Bartosz Adamiczka

approach enables not only a faster selection of the appro-

priate tool but also better alignment with individual design

requirements.

The classication of tools into ve main application

areas enables easier identication and selection of the ap-

propriate tool for a given task without the need to anal-

yse each tool individually. These areas also highlight the

characteristic possibilities of applying AI tools from each

group within the context of using inventive methods. At

the same time, similar to inventive methods, these tools

are not strictly tied to architecture, indicating the need for

an interdisciplinary approach to their use – not as prob-

lem-solving tools but rather as instruments that support the

process of creative problem-solving.

It should be noted that the developed system allows for

adjusting the selection of AI tools to the creator’s pref-

erences regarding their functionality and characteristics.

Furthermore, it also enables decision-making based on the

stage of the creative problem-solving process and the spe-

cic inventive method, which is related to the work with

the sought-after AI tool.

Summary

The study provides new insights into applying AI tools

in the creative problem-solving process, particularly in

the context of inventive methods. Through the conducted

research, a tool enabling the practical use of AI in archi-

tectural design has been developed. An open classication

system for AI tools was established based on their avail-

ability, popularity, and intuitiveness. This lls a gap in the

existing literature, oering a tool that allows for compar-

ing and selecting AI solutions that support designers at

various stages of the creative process. Five application

areas for AI tools were dened, signicantly simplifying

their identication and matching to specic tasks. This ap-

proach eliminates the need for a detailed analysis of each

tool individually. The classication by application areas

and their parameterization allows for quick comparison

and selection of the most suitable solutions, while the de-

veloped tabular summaries facilitate decision-making. As

a result, all the set goals were achieved, creating an AI tool

system tailored to the preferences of creators, considering

the context of the stages of the creative problem-solving

process and inventive methods.

The study focused on widely available and intuitive

tools, excluding more advanced, specialized solutions.

The direct eectiveness of the tools was not assessed, nor

was there any indication of direct integration with specic

inventive methods, leaving room for further verication of

their practical application. The dynamic development of

AI means that the list of tools will require regular updates.

The results of the study and research limitations indi-

cate the need for further verication of the possibilities

and eectiveness of integrating AI tools with inventive

methods, including within practical design applications.

Fig. 2. Diagram and steps of the creative problem-solving process path model

as a strategy for architectural form design

(source: Adamiczka 2022, 109)

Il. 2. Diagram i kroki modelu ścieżki procesu twórczego rozwiązywania problemów

jako strategii projektowania formy architektonicznej

(źródło: Adamiczka 2022, 109)

Systematics of AI tools supporting creative problem-solving and inventive methods in architectural design 115

A B C D

Step of the creative

problem-solving model path

The subject of the step

and the role of the designer

Application Area Indicating example use cases at different stages of the process

Step 1:

Environmental analysis

Subject: Problem identification,

open formulation of the problem.

Designer: Understanding and defining

the problem (issue / circumstances).

Text oriented tools

Generating initial definitions of the problem, creating questions supporting analysis, transforming collected data into understandable

text or reports.

Image oriented tools Visualizing the problem environment, creating mind maps, generating images representing the issue.

Video oriented tools Creating dynamic visualizations of the problem or scenarios, documenting the problem environment using edited video recordings.

Sound oriented tools Recording or sound analysis related to the context of the problem.

Automation oriented tools Automating the collection and analysis of environmental data, generating data visualizations and process maps.

Step 2:

Discovery of aims

Subject: Defining the scope

of the problem

Designer: Formulating the problem

as a question / set of questions

Text oriented tools Generating creative questions supporting problem exploration, creating open-ended queries.

Image oriented tools Creating visual representations of questions or challenges, which helps in better understanding them.

Video oriented tools Generating a video presenting questions in the context of the problem, visualizing challenges in a narrative way.

Sound oriented tools Simulating challenges in an auditory form.

Automation oriented tools Automatically organizing questions in the form of presentations or diagrams, organizing conclusions gathered in step 1.

Step 3:

Discovery of facts

Subject: Information gathering,

thorough problem recognition

Designer: Discovering the relationship

between the problem

and known solutions

Text oriented tools Creating timelines based on textual data, generating summaries of complex information, historical analysis, and forecasting the future.

Image oriented tools Generating diagrams, charts, and historical maps presenting data in an visually appealing format.

Video oriented tools Creating dynamic timeline visualizations, presenting trends in a video format.

Sound oriented tools Analyzing audio data on a timeline, identifying trends in the acoustic environment.

Automation oriented tools Automated analysis of historical data, visualizing trends and their relationships over time, identifying key moments.

Step 4:

Discovery of the problem

Subject: Defining the problem

Designer: Redefining the problem

in the context of the gathered

information

Text oriented tools Creating a new problem definition based on data analysis, paraphrasing, and refining formulations.

Image oriented tools Updating visual representations of the problem based on the collected information.

Video oriented tools Creating video presentations of the new problem definition based on the collected information.

Sound oriented tools Adding narration or sound effects to the redefined problem, helping to illustrate changes in the acoustic environment.

Automation oriented tools Generating reports that redefine the problem, visualizing data updated based on new information.

Step 5:

Clarification of assumptions

Subject: Planning actions leading

to the solution of the problem

Designer: Choosing the technique

for creating forms representing

solution ideas

Text oriented tools Generating lists of tools and techniques, organizing work schedules.

Image oriented tools Creating graphical cheat sheets and visual instructions.

Video oriented tools Preparing video tutorials supporting the use of techniques and tools.

Sound oriented tools Recording audio instructions and guidelines.

Automation oriented tools Automating schedule creation, organizing work based on established priorities.

Table 2. Possibilities of using AI tools in the context of the steps in the model path of the creative problem-solving process (elaborated by B. Adamiczka)

Tabela 2. Możliwości wykorzystania narzędzi AI w kontekście kroków modelu ścieżki procesu twórczego rozwiązywania problemów (oprac. B. Adamiczka)

116 Bartosz Adamiczka

A B C D

Step of the creative

problem-solving model path

The subject of the step and the role of the

designer

Application Area Indicating example use cases at different stages of the process

Step 6:

Discovery of ideas

Subject: Creating solutions

to the problem

Designer: Generating the greatest

possible number of solution ideas

Text oriented tools Generating textual descriptions of ideas and expanding their forms.

Image oriented tools Creating visual forms of ideas and their transformations.

Video oriented tools Generating prototype video visualizations presenting the concepts of ideas.

Sound oriented tools

Creating audio interpretations of ideas, e.g.,

sound effects illustrating different concepts.

Automation oriented tools Automatic sorting and organization of work results.

Step 7:

Discovery of solutions

Subject: Evaluation and selection

of possible solutions

Designer: Defining evaluation criteria

and selecting the most promising

solutions

Text oriented tools

Assistance in formulating evaluation criteria,

organizing results of idea assessments.

Image oriented tools Visualization of evaluation results, e.g., in the form of tables or charts.

Video oriented tools Preparing video presentations comparing different ideas and their assessments.

Sound oriented tools

Adding audio explanations or opinions

to the evaluation results.

Automation oriented tools

Automatic processing of evaluation results

and generating summary reports.

Step 8:

Discovery of acceptance

Subject: Implementation of selected ideas

Designer: Implementing the selected

solution and realizing it in the project

Text oriented tools

Creating implementation reports, contextual analyses;

documenting changes and adapting the implemented solution to the project conditions.

Image oriented tools

Creating prototype graphics, visualizing changes and integrating new concepts

with the existing project environment.

Video oriented tools

Creating video presentations of prototypes, generating animations

or videos showing how the implemented solution functions in practice.

Sound oriented tools Implementation narration and audio presentation.

Automation oriented tools Organizing team work, managing tasks, and monitoring progress.

Step 9:

Control

Subject: Verification of whether the goal

has been achieved

Designer: Comparison of the implement-

ed solution with the current one and with

the assumptions

Text oriented tools Creating text-based comparisons of implemented solutions with initial assumptions.

Image oriented tools

Creating comparative graphics of „before and after”,

showing the differences between the implementation and the initial assumptions.

Video oriented tools

Preparing video materials comparing the effects of the implementation

with earlier visions and project concepts.

Sound oriented tools

Recording audio summaries of the project,

creating narrative reports in the form of podcasts or recordings.

Automation oriented tools

Generating reports and outcome summaries, comparing them with the assumptions,

which streamlines the analysis of the effectiveness of the implementation.

Table 1 continued. Possibilities of using AI tools in the context of the steps in the model path of the creative problem-solving process (elaborated by B. Adamiczka)

Tabela 2 cd. Możliwości wykorzystania narzędzi AI w kontekście kroków modelu ścieżki procesu twórczego rozwiązywania problemów (oprac. B. Adamiczka)

Systematics of AI tools supporting creative problem-solving and inventive methods in architectural design 117

An important aspect is also their inclusion in architectural

education, which will enable future designers to eectively

and consciously use AI tools in the creative process.

The developed systematics supports designers by fa-

cilitating the conscious selection and choice of AI tools

– both according to their preferences and in the context

of the stages of the creative problem-solving process and

the nature of the methods being used. The study opens

new perspectives for architects and creators, enabling the

implementation of innovative technologies in the creative

process to support creativity rather than a replacement.

Translated by

Bartosz Adamiczka

References

Adamiczka, Bartosz. “Proces twórczego rozwiązywania problemów

jako strategia projektowania formy architektonicznej w kontekście

zja wiska percepcji.” PhD diss., Politechnika Wrocławska, 2022.

Almaz, Amira Fawzy, Elsayed Abd El-azim El-Agouz, Mohab Taher Ab-

delfatah, and Islam Rafaat Mohamed. “The Future Role of Articial

Intelligence (AI) Design’s Integration into Architectural and Interior

Design Education Is to Improve Eciency, Sustainability, and Cre-

ativity.” Civil Engineering and Architecture 12, no. 3 (May 2024):

1749–72. https://doi.org/10.13189/cea.2024.120336.

Arnheim, Rudolf. Sztuka i percepcja wzrokowa: Psychologia twórczego

oka. Ocyna, 2013.

As, Imdat, Siddharth Pal, and Prithwish Basu. “Articial Intelligence in

Architecture: Generating Conceptual Design via Deep Learning.”

International Journal of Architectural Computing 16, no. 4 (Decem-

ber 2018): 306–27. https://doi.org/10.1177/1478077118800982.

Brandl, Robert, and Ellis Cai. “ChatGPT Statistics and User Numbers

2024 – OpenAI Chatbot.” Tooltester (blog), Published May 29,

2024. Accessed November 15, 2024. https://www.tooltester.com/en/

blog/ chatgpt-statistics/.

Broma, Tomasz. “Model warstw materialnych jako metoda analizy i pro-

jektowania kształtu formy strukturalnej obiektu architektoniczne-

go.” PhD diss. Politechnika Wrocławska, 2023.

Chaillou, Stanislas. “AI and Architecture: An Experimental Perspective.”

In The Routledge Companion to Articial Intelligence in Architec-

ture, edited by Imdat As and Prithwish Basu. Routledge, 2021.

Gage, Mark Foster. “Architecture That Challenges Your Concept of Rea-

lity.” TEDx Talks, Video, 13 min, 29 sec. Published February 24,

2017. Accessed November 15, 2024. https://www.youtube.com/

watch?v=7v5hmQt57lc.

Hegazy, Muhammad, and Ahmed Saleh. “Evolution of AI Role in Ar-

chitectural Design: Between Parametric Exploration and Machine

Hallucination.” MSA Engineering Journal 2, no. 2 (March 2023):

262–88. https://doi.org/10.21608/msaeng.2023.291873.

Jaruga-Rozdolska, Anna. “Articial Intelligence as Part of Future Prac-

tices in the Architect’s Work: MidJourney Generative Tool as Part of

a Process of Creating an Architectural Form.” Architectus, no. 3 (71)

(2022): 95–104. https://doi.org/10.37190/arc220310.

Kwiatkowska, Ada. “Architectural Formation: The Imperfect Structur-

al Stability and Perfect Instability of the Creation’s Language.” In

Dening the Architectural Space: Rationalistic or Intuitive Way to

Architecture. Vol. 2, edited by Tomasz Kozłowski, Cracow Univer-

sity of Technology, 2018.

Kwiatkowska, Ada. “Mind-Games: Innovative Architectural Design in

the Digital Age.” In Proceedings of 2007 International Conference

on Architectural Education, China Central Academy of Fine Arts,

China Architecture and Building Press, 2007.

Lynch, Kevin. Obraz miasta. Archivolta, 2011.

Mrosla, Laura, Volker Koch, and Petra von Both. “Quo vadis AI in Ar-

chitecture? Survey of the current possibilities of AI in the architec-

tural practice.” In Proceedings of 37 eCAADe and XXIII SIGraDi

Joint Conference, “Architecture in the Age of the 4th Industrial

Revolution”, Porto 2019, edited by José Pedro Sousa, Gonçalo Cas-

tro Henriques and João Pedro Xavier. Blucher 2019. https://doi.

org/10.5151/proceedings-ecaadesigradi2019_302.

Nigam, Aditya, Rhitvik Pasricha, Tarishi Singh, and Prathamesh Chu-

ri. “A Systematic Review on AI-Based Proctoring Systems: Past,

Present and Future.” Education and Information Technologies 26

(September 2021): 6421–45. https://doi.org/10.1007/s10639-021-

10597-x.

Ortega-Bolaños, Ricardo, Joshua Bernal-Salcedo, Mariana Germán Or-

tiz, Julian Galeano Sarmiento, Gonzalo A. Ruz, and Reinel Taba-

res-Soto. “Applying the Ethics of AI: A Systematic Review of Tools

for Developing and Assessing AI-Based Systems.” Articial Intel-

ligence Review 57, art. 110 (April 2024). https://doi.org/10.1007/

s10462-024-10740-3.

Proctor, Tony. Twórcze rozwiązywanie problemów. Podręcznik dla mene-

dżerów. Gdańskie Wydawnictwo Psychologiczne, 2002.

Said, Ahmad Nur Maulana, Diyah Daulah Auliyah Badar, Nurul Cahya -

ni Hamzah, Muhammad Jamil Syah, and Muh Hamka Al Farizi.

“Implementation of Articial Intelligence in the World of Architec-

ture and Its Inuence on the Profession of Architecture.” Prosiding

Seminar Nasional UNM Ke-62 2023 1 (July 2023): 620–26. https://

doi.org/10.59562/semnasdies.v1i1.1117.

Sarkar, Sujan. “AI Industry Analysis: 50 Most Visited AI Tools and

Their 24B+ Trac Behavior – WriterBuddy.” Writerbuddy (blog).

Published October 17, 2023. Accessed November 15, 2024. https://

writerbuddy.ai/blog/ai-industry-analysis.

Sarkar, Sujan. “Top 50 AI Companies of 2024: Funding, Valuation &

Trends.” Writerbuddy (blog). Published 2024. Accessed November

15, 2024. https://writerbuddy.ai/blog/top-50-ai.

Singh, Shubham. “ChatGPT Statistics (NOV. 2024): Number of Users

& Queries.” DemandSage (blog). Published November 3, 2024. Ac -

cessed November 15, 2024. https://www.demandsage.com/chat gpt-

statistics/.

Tarczewski, Romuald. Konstruowanie architektury. Uwagi o materia-

lizacji formy architektonicznej. Ocyna Wydawnicza Politechniki

Wrocławskiej, 2019.

Tellios, Anastasios, Panagiota Koulali, and Kalliopi Valsamidou. “De-

signing Tomorrow: AI and the Future of Architectural Design Pro-

cess.” Interdisciplinary Journal of Architecture and Built Environ-

ment 27 (October 2023): 22–5. https://doi.org/10.37199/f40002703.

Streszczenie

Systematyka narzędzi AI wspierających twórcze rozwiązywanie problemów i metody inwentyczne w projektowaniu architektonicznym

Dynamiczny rozwój narzędzi opartych na sztucznej inteligencji (AI) istotnie wpływa na współczesne procesy projektowe, w tym projektowanie

architektoniczne. Narzędzia te otwierają nowe perspektywy dla twórców, wspierając ich na różnych etapach pracy. AI w architekturze staje się kluczo-

wym elementem wspomagającym projektantów w rozwiązywaniu problemów, jednak nadal brakuje systematycznego ujęcia i klasykacji związanych

z nią narzędzi pod kątem twórczego rozwiązywania problemów, metod inwentycznych oraz podejścia intuicyjnego i myślenia metaforycznego, które

jest istotne w procesie kreacji.

118 Bartosz Adamiczka

Tematem artykułu są możliwości zastosowania powszechnie dostępnych narzędzi sztucznej inteligencji jako wsparcia metod inwentycznych w pro-

cesie twórczego rozwiązywania problemów. Pierwszym celem autora jest usystematyzowanie narzędzi AI pod kątem ich potencjalnych zastosowań

w kontekście metod inwentycznych. Drugim – identykacja sposobów wykorzystania tych narzędzi na poszczególnych etapach twórczego rozwiązy-

wania problemów.

Analiza została oparta na danych pozyskanych metodą desk research oraz klasykacji narzędzi według kryteriów popularności i funkcjonalności.

W ramach badania narzędzia pogrupowano w pięć obszarów zastosowań, odnosząc je do kolejnych etapów procesu twórczego rozwiązywania pro-

blemów.

W toku badań zastosowano metody analizy literatury, obserwacji, dedukcji i syntezy, co pozwoliło na kompleksową werykację pozyskanych

danych. Dodatkowo, proces badawczy oparto na metodach inwentycznych, takich jak: analiza literatury przedmiotu, szukanie związków, odpowied-

niość między teoriami, wycieczka przykładowa, mapy myślowe oraz dedukcja – poprzedzonych metodą intuicyjną, która stanowiła impuls dla działań

koncepcyjnych na każdym etapie badań.

Wynikiem pracy jest otwarta systematyka narzędzi AI umożliwiająca projektantom porównanie i wybór narzędzi najlepiej odpowiadających ich po-

trzebom i preferencjom. Klasykacja narzędzi według obszarów zastosowań umożliwia wybór narzędzia ze względu na charakter wykorzystywanych

metod inwentycznych oraz etap procesu twórczego rozwiązywania problemów. Opracowane zestawienia tabelaryczne pozwalają na świadomą i prostą

identykację narzędzi odpowiednich do poszczególnych zadań. Wyniki badań wskazują na potrzebę dalszej werykacji praktycznego zastosowania

narzędzi AI oraz ich uwzględnienia w edukacji architektonicznej, co pozwoli na pełniejsze wykorzystanie potencjału tych technologii w procesie

projektowym.

Słowa kluczowe: projektowanie architektoniczne, AI, metody inwentyczne, twórcze rozwiązywanie problemów