Available via license: CC BY
Content may be subject to copyright.
Towards Artificial Intelligence Serving as an
Inspiring Co-Creation Partner
Kevin German1, Marco Limm1, Matthias Wölfel2,*, Silke Helmerdig1
1School of Design, Pforzheim University, Pforzheim, Germany
2Faculty of Computer Science and Business Information Systems, Karlsruhe University of Applied Sciences,
Karlsruhe, Germany
Abstract
The promise of artificial intelligence (AI), in particular its latest developments in deep learning, has been
influencing all kinds of disciplines such as engineering, business, agriculture, and humanities. More recently
it also includes disciplines that were “reserved” to humans such as art and design. While there is a strong
debate going on if creativity is profoundly human, we want to investigate if creativity can be enhanced by
AI—not replaced. To be an inspiring co-creation partner by suggesting unexpected design variations and by
learning the designer’s taste. To do so we adopted AI algorithms, which can be trained by a small sample set
of shapes of a given object, to propose novel shapes. The evaluation of our proposed methods revealed that
it can be used by trained designers as well as non-designers to support the design process in different phases
and that it could lead to novel designs not intended/foreseen. Besides the potentials of AI, we also point out
and discuss moral threads caused by the latest developments in AI with respect to the creative sector.
Received on 18 February 2019; accepted on 30 March 2019; published on 26 April 2019
Keywords: inspirational AI, human-machine co-design, artificial neural network, genetic algorithm, design process
Copyright © 2019 Kevin German et al., licensed to EAI. This is an open access article distributed under the terms
of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/3.0/), which permits unlimited
use, distribution and reproduction in any medium so long as the original work is properly cited.
doi:10.4108/eai.26-4-2019.162609
1. Introduction
The promises of AI assisted creation is “a world where
creativity is highly accessible, through systems that
empower us to create from new perspectives and raise
the collective human potential” as Roelof Pieters and
Samim Winiger pointed out. Recent developments in
artificial intelligence (AI) have demonstrated that they
are indeed capable to do things which in the past
were restricted to humans. Artificial neural networks
(ANN) and genetic algorithms (GA) are tools to make
work easier for humans, for example through automatic
speech translations (for instance simultaneous lecture
translation has been demonstrated feasible already
in 2008 by Kolss et al. [12]) or are even able to
come up with solutions humans would never come
up with effortlessly, see for instance the design of
an “evolved antenna” using evolutionary algorithms
published by Hornby et al. already in 2006 [8]. With
further technological developments, of such processes
there is a gradual transfer of competence from human
beings to technical devices, namely, they serve as [24]:
1. tools: transfer of mechanics (material) from the
human being to the device
2. machines: transfer of energy from the human
being to the device
3. automatic machines1: transfer of information from
the human being to the device
4. assistants: transfer of decisions from the human
being to the device
We want to exemplify this concept with the field of
mobility:
1. bicycle: feet are replaced by wheels
2. motor vehicle: propulsion is replaced by engine
3. self-driving rail vehicle: control is replaced by
sensors and signal processing
1which is called Automat or automate in other languages such as
German or French respectively
1
EAI Endorsed Transactions
on Creative Technologies Research Article
EAI Endorsed Transactions on
Creative Technologies
01 2019 - 04 2019 | Volume 6 | Issue 19 | e3
Corresponding author. Email: Matthias@wolfel.de
**
K. German et al.
4. autonomous vehicle: route planning or search
for a parking space are replaced by artificial
intelligence
Similarly, we can give an example from the field of art
and design:
1. potter’s kick-wheel: a tool used in the shaping of
round ceramic ware driven by kicking a fly-wheel
into motion
2. potter’s electric-wheel: the kicking of the fly-
wheel is replaced by a motor
3. construction & 3D printing: the object is con-
structed with a CAD-software according to given
parameters and 3D printed
4. generated & 3D printing: the object is generated
by an optimization process given particular
constraints and 3D printed
In the coming years we are in the process of moving
from Step 3. to Step 4. which raises—as it was the case
from moving from Step 1. to Step 2. as well as from Step
2. to Step 3.—discussions, rejections, ethical issues (see
Section 5), up to fears.
In the literature, some approaches to use AI in
the creative sector have been presented. We review
those approaches in the following section. Because the
already introduced approaches are not available or were
not fulfilling our requirements it was necessary to adopt
given methods to intervene in the design process; either
partially or in total. The investigated algorithms include
genetic algorithms and different versions of ANNs
namely convolutional neural networks, generative
adversarial networks, and variational autoencoder. The
developed algorithms can semi- or fully-automate the
research, brainstorming and concept phase of the
design process.
To evaluate and compare our different proposed
approaches the entire development process was com-
pleted until the finished product for each approach. The
approaches have been introduced within the School of
Design at Pforzheim University, Germany and show-
cased to visitors of the Salone del Mobile in Milan, Italy,
the Dutch Design Week in Eindhoven, Netherlands and
the VDID Congress in Stuttgart, Germany. On these
occasions, we were able to demonstrate that our pro-
posed methods can be used by trained designers as well
as non-designers to design semi-complex shapes with
minimal user feedback.
2. Related Work
The idea of using algorithms to support the creation
process is well established and frequently referred
to as generative design or procedural generation. It is
used to generate geometric patterns, textures, shapes,
meshes, terrain or plants. The generation processes may
include, but are not limited, to self-organization, swarm
systems, ant colonies, evolutionary systems, fractal
geometry, and generative grammars. McCormack et
al. [15] review some generative design approaches and
discuss how design as a discipline can benefit from
those applications. While older approaches rely on
generative algorithms which are usually realized by
writing program code, ANN changes this process into
data driven procedures. ANN can learn patterns from
(labeled) examples or by reinforcement. Wölfel [25]
points out that there are fundamental differences in
the goals and reasons to use AI in art, design and
cultural heritage: while in the former two AI should
help to foster creativity and inspiration in the latter it
should help to (re)discover or enhance given patterns;
e.g. reconstruct part of an image which has been
damaged over time. To create higher variations some
artists randomly introduce glitches within the ANN.
Due to the complex structure of the ANN these glitches
(assuming that they are introduced at an early layer in
the network) occur on a semantic level which causes
the models to misinterpret the input data in interesting
ways, some of which could be interpreted as glimpses of
autonomous creativity; see for instance the artistic work
‘Mistaken Identity’ by Mario Klingemann.
AI or more precise ANNs has been introduced to sup-
port the creation process more recently. Leading soft-
ware companies in engineering and design have already
included AI-driven generative design paradigms which
let humans input design goals. For instance, Project
Dreamcatcher [2] is an engineering-based generative
design program that takes into account how the forces
will be directed best in the product and defines the
best production method. Autodesk states the benefits of
generative design to [1]:
• explore a wider range of design options
• make impossible designs possible
• optimize for materials and manufacturing meth-
ods
Most popular (at least in the mass media) are prob-
ably different variations of image-to-image translation.
The most prominent example is style transfer—the capa-
bility to transfer the style of one image to draw the con-
tent of another. But mapping an input image to an out-
put image is also possible for a variety of other applica-
tions such as object transfiguration (e.g. horse-to-zebra,
apple-to-orange, season transfer (e.g. summer-to-winter)
or photo enhancement [27]. While some of the just men-
tioned system seems to be toy applications, AI tools are
taking over and gradually automate design processes
which used to be time-consuming manual processes.
Indeed, the most potential for AI in art and design is
seen in its application to tedious, uncreative tasks such
2
EAI Endorsed Transactions on
Creative Technologies
01 2019 - 04 2019 | Volume 6 | Issue 19 | e3
Towards Artificial Intelligence Serving as an Inspiring Co-Creation Partner
as coloring black-and-white images [26]. Cluzel et al.
have proposed an interactive GA to progressively sketch
the desired side-view of a car profile [3]. For this, the
user has taken on the role of a fitness function2through
interaction with the system. The chAIr Project [20] is
a series of four chairs co-designed by AI and human
designers. The project explores a collaborative creative
process between humans and computers. It used a gen-
erative adversarial network (GAN) to propose new chairs
which then have been ‘interpreted’ by trained designers
to resemble a chair. DeepWear [9] is a method using
deep convolutional GANs for clothes design. The GAN
is trained on features of brand clothes and can generate
images that are similar to actual clothes. A human
interprets the generated images and tries to manually
draw the corresponding pattern which is needed to
make the finished product. Li et al. [14] introduced an
ANN for encoding and synthesizing the structure of 3D
shapes which—according to their findings—are effec-
tively characterized by their hierarchical organization.
Marco Kempf and Simon Zimmerman used AI in their
work dubbed ‘DeepWorld’ to generate a compilation of
‘artificial countries’ using data of all existing countries
(around 195) to generate new anthems, flags and other
descriptors. Roman Lipski uses an AI muse (developed
by Florian Dohmann et al.) to foster his/her inspiration.
Because the AI muse is trained only on the artists previ-
ous drawings and fed with the current work in progress
it suggests image variations in line with Romans taste.
Most of the related work is not ready yet to
be used without a thorough understanding of the
technology and is more an engineering approach
using ANNs instead of common technology. Daniel
Wikström [23] mentions that many designers do not yet
know technology well enough and therefore perceive
it as “magic”. But he also explains how an intelligent
assistant is perceived and would have to interact. What
we are aiming for is different: The whole creation
process—not its development—should be applicable to
naive users without any profound understanding of
design or engineering. The user has to only rely on
his/her taste to cherry-pick examples he/she likes in
an iterative process until he/she ends up with the final
design.
3. Co-Creating Shapes with Artificial Intelligence
In this section we want to introduce different
approaches how to use AI as a co-creation partner.
Before we can start to do so, however, we first have to
develop an approach which is able to suggest plausible
shapes of bottles. We chose bottles because they have
simple shapes and thus keep the necessary training
2also referred to as objective function
effort low, but still are able to express unique features
which can easily be recognized; just think of the iconic
contour fluted lines of the Coca-Cola bottle.
3.1. Plausible Shape Representation
A “naive” approach to automatically generate bottle
shapes would be to start by drawing randomly placed
polygons or polylines and optimize it by targeted
selection, mating, recombination and mutation by
optimizing to a particular goal; e.g. has to look like a
bottle. To determine if an image ‘looks like a bottle’ pre-
trained classifiers could be used (see e.g. Krizhevsky et
al. [13]). However, as has been pointed out by German et
al. [4] this “naive” approach is not leading to satisfying
results. The problem here is that not only shapes which
are similar to bottles get a high score, but also shapes
with random patterns are able to get similar scores. This
is a problem, known as adversarial examples [6], and not
uncommon in ANNs.
T
raining Data
Generator
Original
Discriminator
Fake
Figure 1. Flow chart of generative adversarial network and
different instances according to the different steps.
3
EAI Endorsed Transactions on
Creative Technologies
01 2019 - 04 2019 | Volume 6 | Issue 19 | e3
K. German et al.
Therefore, an approach is required which guarantees
that the produced shapes are similar to the shape
of bottles. In 2014 Goodfellow et al. proposed the
special ANN architecture GAN which we have already
mentioned before [5]. The main idea of their proposal
is to use two ANNs that compete with each other. Fig. 1
demonstrates the basic principle and components: The
generator tries to generate data from latent variables
that are as similar as possible to the training data.
The discriminator tries to classify the generated data
according to the original training data. Both networks
play a zero-sum game: As the system progresses, the
generator as well as the discriminator are improving.
This process continues until the discriminator can no
longer distinguish between forgery and original. This is
achieved when the discriminator is only correct in 50%
of the cases.
Since the generator learns to generate data as similar
as possible to the training data, it requires a training
data set that corresponds as closely as possible to the
desired output [5]. In our case we were interested in
generating different variations of shapes resembling
bottles. To train our system we converted 200 images
of bottles into black and white silhouettes (see Fig. 2).
As automatic segmentation did not lead to satisfactory
results the conversation was done by hand. Since the
data volume is small and GANs normally use data
volumes in orders of magnitude of several 1,000 images,
there is a risk of over-adaptation by the GAN [22].
To reduce over-adaptation, data augmentation is used
by automatically generating variations of the available
training data including shearing, enlarging, rotating
and cropping. In order to further reduce the overfitting
of all the methods presented here, a considerable
amount of regularization [17] and dropout layers [21]
were used.
Fig. 3shows that the training loss in the first few
generations quickly approaches zero. This is due to the
fact that the network initially roughly maps the basic
form of the input data. In higher epochs many bottles
of an epoch have similar characteristics. This is a well-
known problem in GAN architectures and is called
mode collapse. The generator limits itself to generating
only a few examples that the discriminator classifies
as original. In the worst case, all images generated
by the generator are almost identical [16]. Although
in our example we see variations the problem is still
visible. Different epochs can be considered to create
more diverse bottles because the point of mode collapse
shifts with each epoch. Although the training data set
only consists of symmetrical bottles, the architecture
is capable of generating asymmetric bottles. This is
interesting because the net is able to generate something
it did not know could be e.g. asymmetric. It is up to
the designer to incorporate these unusual features such
as asymmetrical elements into the product design or to
rate them as a mistake and to correct them manually
based on his/her taste.
Due to the required minimum complexity of the GAN
architectures and the need for sharp high-resolution
images in combination with the low amount of training
data, overfitting inevitably occurs. However, subjective
comparisons with the training data set did not rate the
over-adaption as critical as the majority of the bottles
are unique. Instead of treating the shape as one union
it might be advantageous to separate the shape into
different parts.
3.2. Semantic Shape Representation
The shape of an object can be decomposed into
different features that can be assigned with particular
“meanings” and semantically annotated3. In our
particular application of a bottle the semantic shape
representation can be separated and annotated into:
lid, neck, wall, wall-to-neck transition and bottom4.
The classification was done manually by cutting the
existing 200 images into individual parts. In the future,
however, this step can be automated using image
segmentation.
One conceivable option for creating new shapes of
bottles is the random permutation of the semantic parts
and thus to overcome the limiting characteristics of
the former approach where many generated bottles had
similar characteristics. For this purpose, an ANN is to
be conceptualized, which receives random features and
assembles them to form a new object. The network
learned, in the training phase, the relationship between
the semantic features and the actual bottle. After this
phase, the network is able to merge features seamlessly
and to produce the shape of a consistent bottle. New
permutations of features using the trained ANN are
shown below in Fig. 4. The features were determined
based on a discrete equal distribution. It can be
observed that the features are transferred and combined
successfully.
3.3. Introducing Personal Taste in Shape
Representation
So far we have described the process of how to fully
automatically generate plausible shapes by varying
different features of the bottle. Now it’s time to bring
back the user by having him/her intervene in the design
process: The shape should advance iteratively towards
the taste of the user. For optimization problems in
which a solution approaches an optimum step by step,
3Semantic annotation is the process of attaching additional
information to various concepts to be used by machines.
4In preliminary tests, this division turned out to be the most effective
variant.
4
EAI Endorsed Transactions on
Creative Technologies
01 2019 - 04 2019 | Volume 6 | Issue 19 | e3
Figure 2. Black and white silhouettes of bottles.
Figure 3. Different iterations of the learning process. From left to right, iteration 50, 100, steps of 100 until 1000. Four different
examples are shown for each iteration.
GA has already proven to be an appropriate tool [8],
which is also why a GA was used in this procedure. The
basic idea is that you have a population of objects where
each object is defined by its genes. Each gene represents
a semantic feature, in this case, e.g., the bottleneck.
To transform the genes into visible features, the ANN
of the semantic shape representation is used. Another
valid possibility that was not considered here would be
to use the genes as an input vector for a generator or
variational autoencoder to express the visible shape.
Similar to the biological model, the population
gradually adapts to the environment through selection,
mating, gene recombination and mutation [7]. To
introduce the designer into the automatic algorithm
the random permutations of features have to be
evaluated by the designer instead of a genetic objective
function. Therefore, the designer takes up the position
of the fitness/objective function, similar to the ANN
MobileNet, by sitting in front of the computer and by
evaluating each instance individually; Fig. 5. The basic
idea here is that the population gradually approaches
the taste of the user until his/her ideal bottle is created.
Therefore, each of the 20 individuals in the population
is assigned a fitness value between zero and one by
the user. The higher the fitness value the higher the
probability of survival by an individual. Combined with
the previously mentioned methods such as mutation,
this results in a population which is more precisely
adapted to the taste of the user. To cover a large search
space, the population is initialized using a discrete
equal distribution. Over a couple of iterations the final
optimal shape is found.
3.4. Democratizing Shape Representation
To be able to democratize the design process we have
to vary the proposed approaches so far to be able to do
5
Towards Artificial Intelligence Serving as an Inspiring Co-Creation Partner
EAI Endorsed Transactions on
Creative Technologies
01 2019 - 04 2019 | Volume 6 | Issue 19 | e3
K. German et al.
lid neck
1.
2.
3.
wall transition bottom prediction
Figure 4. Three variations of bottle shapes as generated by merging the decomposed parts as given by the semantic shape
representation approach.
some arithmetic’s; e.g. to calculate the arithmetic mean
of a set of bottles designed by different persons. To do
so we use a variational autoencoder (VAE) [10]. It is an
ANN that learns to produce the same output as input.
A special feature here is that the network topology has
a bottleneck between the input and the output layers.
This bottleneck stores the compressed information as a
vector of real numbers called latent variables (LV). As a
result, the autoencoder must compress information of
the input into the LV and then decompress it after the
bottleneck. The VAE learns to extract the most relevant
information from an input image as LV so that it can
be used to regenerate the output image as correctly as
possible [10].
The basic idea is this: After training, the LV can be
accessed directly through sliders. The trained decoder
would then convert the LV into a corresponding bottle.
This would allow the non-designer with limited design
skills to design an object in a playful way (Fig. 6). A
number of eight LV have delivered satisfactory results
in trials. A smaller number of LV leads to less detailed
and more similar images. More LV, on the other hand,
have not achieved any significant improvement in
quality, but have worsened the user experience due to
more necessary sliders.
It can be seen that by moving individual sliders,
the bottle can be transferred into other forms.
The transformation is done simultaneously with the
slider movement, giving the user direct and intuitive
feedback. A complete disentanglement of the LV could
not be achieved. Consequently, a LV and thus the
corresponding slider can be responsible for several
semantic features of the object.
Because there are vectors behind the bottles, we
can do bottle arithmetic with them [19]. This makes
it possible to calculate the arithmetic mean of a
set of bottles. This allows several individuals to
democratically design a bottle by first creating a bottle
for each individual using the sliders and then averaging
all created bottles. There are two main pillars of
6
EAI Endorsed Transactions on
Creative Technologies
01 2019 - 04 2019 | Volume 6 | Issue 19 | e3
Create Start
Population
Selection
Result
Evaluation by User
Recombination
Mutation
Figure 5. Flow chart of the genetic algorithm and different
instances according to the different steps.
Figure 6. Transforming the bottles using eight parameters. Each
slider corresponds to one latent variables.
democratic design. First, anyone can design objects now
even without design skills and secondly, the taste of
each individual can equally influence the final product.
4. Results, Evaluation & Limitations
Using the plausible shape representation (Section 3.1)
method, it was shown that parts of the design process
can be partially automated and thus speed up using
ANNs. This architecture typically provides a good
image quality. However, the algorithm does not allow
direct access by the designer, so the output is heavily
dependent on the training data. For instance, to
specifically design a classic beer bottle, the designer
would have to explicitly look for the shape in the output
or to use only beer bottles as training data. Although
novel bottle shapes are created, these usually do not
deviate much from the training data set. For the design
process, the user has received some suggestions from
the algorithms and has decided on one of these in
several iterations; see Fig. 7a and 8a.
Using the semantic shape representation (Section 3.2),
new bottles could also be created automatically. In
comparison to the previous method, these objects are
more diverse and creative looking; see Fig. 7b and 8b.
At present, there must be a database of the specific
objects and their associated features available, which is
not ideal. The image quality is slightly worse than in
the plausible shape representation, but still at a very
high level. In addition, as with the plausible shape
representation, the problem is that the suggestions are
not adapted to the user.
To tackle the latter problem, personal taste (Sec-
tion 3.3) was introduced into the semantic shape repre-
sentation. The bottles successfully adapted to the taste
of the user through evolution; see Fig. 7c and 8c. A
selection from a large amount of output data as in
the last two algorithms is thereby eliminated (apart
from the fitness score evaluation). In our opinion this
is one of the most promising ways to liberate design
processes in the future because designing personalized
objects according to his/her taste becomes possible for
everybody. The algorithm also adapts through the direct
feedback dynamically to changes in the user’s taste,
for instance, during a lifetime. Since the architecture is
based on the semantic shape representation, the image
quality is at the same level and a database of associated
features is also needed.
Through the democratizing shape representation (Sec-
tion 3.4) method, see Fig. 7d and 8d, collectives can
design objects together. With the introduction of vari-
able parameters (sliders), every human being is able to
design things, whether talented or not. This bypasses
the designer and allows the end-user to take on the
role of a designer directly. Secondly, the opinion of each
individual can be incorporated into a final product.
7
Towards Artificial Intelligence Serving as an Inspiring Co-Creation Partner
EAI Endorsed Transactions on
Creative Technologies
01 2019 - 04 2019 | Volume 6 | Issue 19 | e3
K. German et al.
Figure 7. 3D print of generated bottles using a. plausible shape representation, b. semantic shape representation, c. personal taste
in shape representation, and d. democratized shape representation
There’s no need for a central design instance anymore.
The zeitgeist of the collective can (anonymously) create
something together, on which the majority can agree
on. Also, the manual sketches of the concept phase
were eliminated. Within a few seconds, countless new
variants could be created, for which otherwise individ-
ual manual sketches would be needed. However, the
image quality and diversity are worse compared to the
previous algorithms.
In Table 1we compare the different approaches
according to the parameters described next:
•Affordance (in data preparation) describes how
much time has to be spent to prepare the data to
train the ANN.
•Automation describes how much the process is
automated and how much amount has to be done
by the designer.
8
EAI Endorsed Transactions on
Creative Technologies
01 2019 - 04 2019 | Volume 6 | Issue 19 | e3
Figure 8. Rendering of generated bottles using a. plausible shape representation, b. semantic shape representation, c. personal taste
in shape representation, and d. democratized shape representation
Table 1. Comparison of the different methods presented here.
plausible semantic personal democratic
shape shape taste approach
Affordance medium high medium medium
Automation full full semi semi
Shape quality very good good good medium
Creativity medium very good very good medium
Personalization low low high medium
•Shape quality describes the subjective quality
of the shape including detail density, image
sharpness, resolution and number of image
artifacts.
9
Towards Artificial Intelligence Serving as an Inspiring Co-Creation Partner
EAI Endorsed Transactions on
Creative Technologies
01 2019 - 04 2019 | Volume 6 | Issue 19 | e3
K. German et al.
•Creativity describes to what extent the automati-
cally generated results have a creative or inspiring
effect on the designer.
•Personalization describes how much individuality
is kept in the design process and how much of the
personal taste is represented in the outcome.
As previously mentioned all variants shown here
were trained with a well-defined data set consisting of
200 relatively simple 2D images. This procedure was
sufficient to analyze the process. If the same procedures
can be applied to more complex shapes and higher
dimensionality is unclear because these variants might
encounter additional problems. A possible solution in
the future would be the use of voxels or a polygon
mesh, which allows a 3D representation. However,
experience shows that the necessary amount of training
data increases with increasing complexity. A manually
created data set is therefore no longer a valid option.
With automatically created 2D data sets, e.g. by web
scraping, this leads to problems because these images
often have a lack of quality for this application, for
instance by having other objects in the image or through
image artifacts (which is however desired for image
classifications due to better generalization). For 3D
objects, this is not to this extent the case, e.g. CAD
files in most cases only depict the desired object. To
get this data, there are already large databases that have
high quality [11]. Because CAD is an industry-standard,
companies can also use their existing data-sets. The
disadvantage of the increased complexity due to the 3D
representation can potentially be partly compensated
by the high quality and quantity of the training data.
5. Moral Threads
For decades, AI has fostered (often false) future
visions ranging from transhumanist utopia to “world
run by machines” dystopia. Artists and designers
explore solutions concerning the semiotic, the aesthetic
and the dynamic realm, as well as confronting
corporate, industrial, cultural and political aspects.
The relationship between the artist and the artwork is
directly connected through their intentions, although
currently mediated by third-parties and media tools.
Understanding ethical and social implications, in
particular possible threads, of AI assisted creation is
becoming a pressing need and include:
•Wrong Expectations: Only "working examples"
are demonstrated in the media, therefore wrong
expectations are raised. A lot of content claiming
to be AI has indeed been produced by methods
not containing AI (not only in the creative
community). Wrong expectations are leading to
worries about: design AI tools that replace us or
design AI tools that shape us after we shape them
(adapted from Marshall McLuhan)
•Data Bias: Al systems are sensitive to bias. As a
consequence, the AI is not being a neutral tool,
but has pre-decoded preferences. Bias relevant in
creative AI systems are: algorithmic bias occurs
when a computer system reflects the implicit
values of the humans who created it; data bias
occurs when your samples aren’t representative of
your population of interest; prejudice bias results
out of cultural influences or stereotypes.
•Authorship: The authorship of AI generated
content has not been clarified. For instance, is the
authorship of a novel song composed by an AI
trained exclusively on songs by Johann Sebastian
Bach belonging to the AI, the developer, or Bach?
See e.g. [18] for a more detailed discussion.
•Replacement: Do we still need a designer in times
of AI and automation? Not only is this the first
question that crosses the minds of non-designers,
but it is an even more important question for the
design world. Designers are not the only ones to
feel the thread of AI. For instance, translators are
concerned that they could be replaced through
machine translation and truck drivers fear to lose
their jobs because of autonomous driving.
6. Conclusion & Outlook
In this work, we set out to prove that most of the
creation process could be automated or at least semi-
automated and that a workflow from the first sketches
to the final product can be supported by AI. This
became possible by generating design proposals from
different algorithms including ANN and GA. This
drastically accelerated the creation process and saved
tedious labor time. The potential of AI in creativity
has just been started to be explored: AI is shifting
the creativity process from crafting to generating and
selecting; AI has a high potential in the creative sector it
can lower the time between intention and realization; it
can potentially lead to the democratization of creativity.
We chose a simple object—a bottle—to prove our con-
cept. Any other object could, in principle, be designed
the same way. It should be also possible to extend our
proposed approach to include a third dimension. More
complex shapes and higher dimensionality, however,
raises complexity and therefore more data and other
solutions might need to be introduced.
We live in an era of accelerating technological
progress which is already influencing our daily lives.
We cannot ignore technological developments and
pretend these changes are not happening. Instead,
we should embrace the development—but also reflect
10
EAI Endorsed Transactions on
Creative Technologies
01 2019 - 04 2019 | Volume 6 | Issue 19 | e3
its impact—and see it as a new set of opportunities
for us to explore and prosper. Widespread misuse
(threads) can limit the social acceptance and requires
an AI literacy—just like digital literacy—for everybody.
We have to reflect on what makes us human and
remember that we are still the ones who are conceiving
something that we think of as beautiful and therefore
value it. “Successful designs are not necessarily ‘made’:
new functionality may ‘evolve’ through the use and
interpretation of artifacts by an audience” [15]. There
are many examples today where AI has influenced the
creative process letting the designer cherry-pick and
approve adjustments based on the proposed variations.
Let us start exploring these possibilities today and see
where they can take us.
References
[1] AutodeskResearch (2019) Generative design.
URL https://www.autodesk.com/solutions/
generative-design. Accessed: 2019-11-27.
[2] AutodeskResearch (2019) Project dreamcatcher.
URL https://autodeskresearch.com/projects/
dreamcatcher. Accessed: 2019-11-27.
[3] Cluzel, F.,Yannou, B. and Dihlmann, M. (2012)
Using evolutionary design to interactively sketch
car silhouettes and stimulate designer’s creativity.
Engineering Applications of Artificial Intelligence 25(7):
1413–1424.
[4] German, K.,Limm, M.,Wölfel, M. and Helmerdig,
S. (2019) Co-designing object shapes with artificial
intelligence. In Proceedings of ArtsIT.
[5] Goodfellow, I.,Pouget-Abadie, J.,Mirza, M.,Xu,
B.,Warde-Farley, D.,Ozair, S.,Courville, A. et al.
(2014) Generative adversarial nets. In Advances in neural
information processing systems: 2672–2680.
[6] Goodfellow, I.J.,Shlens, J. and Szegedy, C. (2015)
Explaining and harnessing adversarial examples.
[7] Gupta, D. and Ghafir, S. (2012) An overview of
methods maintaining diversity in genetic algorithms.
International journal of emerging technology and advanced
engineering 2(5): 56–60.
[8] Hornby, G.,Globus, A.,Linden, D. and Lohn, J.
(2006) Automated antenna design with evolutionary
algorithms. In American Institute of Aeronautics and
Astronautics Conference on Space, San Jose, CA: 19–21.
[9] Kato, N.,Osone, H.,Sato, D.,Muramatsu, N. and
Ochiai, Y. (2018) Deepwear: a case study of collaborative
design between human and artificial intelligence. In
Twelfth International Conference on Tangible, Embedded,
and Embodied Interaction (ACM): 529–536.
[10] Kingma, D.P. and Welling, M. (2013) Auto-encoding
variational bayes. arXiv preprint arXiv:1312.6114 .
[11] Koch, S.,Matveev, A.,Jiang, Z.,Williams, F.,Artemov,
A.,Burnaev, E.,Alexa, M. et al. (2019) ABC: A big
CAD model dataset for geometric deep learning. In
Proceedings of the IEEE Conference on Computer Vision and
Pattern Recognition: 9601–9611.
[12] Kolss, M.,Wölfel, M.,Kraft, F.,Niehues, J.,Paulik,
M. and Waibel, A. (2008) Simultaneous german-english
lecture translation. In International Workshop on Spoken
Language Translation.
[13] Krizhevsky, A.,Sutskever, I. and Hinton, G.E. (2012)
Imagenet classification with deep convolutional neural
networks. In Advances in neural information processing
systems: 1097–1105.
[14] Li, J.,Xu, K.,Chaudhuri, S.,Yumer, E.,Zhang, H.
and Guibas, L. (2017) Grass: Generative recursive
autoencoders for shape structures. Transactions on
Graphics (TOG) 36(4): 52.
[15] McCormack, J.,Dorin, A.,Innocent, T. et al. (2004)
Generative design: a paradigm for design research.
Proceedings of Futureground, Design Research Society,
Melbourne .
[16] Metz, L.,Poole, B.,Pfau, D. and Sohl-Dickstein, J.
(2017) Unrolled generative adversarial networks. In
Proceedings of 5th International Conference on Learning
Representations.
[17] Ng, A.Y. (2004) Feature selection, l 1 vs. l 2
regularization, and rotational invariance. In Proceedings
of the twenty-first international conference on Machine
learning (ACM): 78.
[18] Rack, F. and Vettermann, O. (2019) KI-Kunst und
Urheberrecht – die Maschine als Schöpferin? URL http:
//tlmd.in/a/3353. Accessed: 2019-11-27.
[19] Radford, A.,Metz, L. and Chintala, S. (2015) Unsu-
pervised representation learning with deep convolu-
tional generative adversarial networks. arXiv preprint
arXiv:1511.06434 .
[20] Schmitt, P. and Weiss, S. (2018) The chair project—four
classics. URL https://philippschmitt.com/work/
chair. Accessed: 2019-11-27.
[21] Srivastava, N.,Hinton, G.,Krizhevsky, A.,Sutskever,
I. and Salakhutdinov, R. (2014) Dropout: a simple way
to prevent neural networks from overfitting. The journal
of machine learning research 15(1): 1929–1958.
[22] Wang, J. and Perez, L. (2017) The effectiveness of
data augmentation in image classification using deep
learning. Convolutional Neural Networks Vis. Recognit .
[23] Wikström, D. (2018) Me, myself, and AI: Case study:
human-machine co-creation explored in design.
[24] Wölfel, M. (2018) Der smarte Assistent. In: Smartphone-
Ästhetik: zur Philosophie und Gestaltung mobiler Medien.
Ruf, O. (editor). Transcript : 269–288.
[25] Wölfel, M. (2020) Bringing cultural heritage to life
through the lens of artificial intelligence. In Proceedings
of HCI International.
[26] Zhang, G.,Qu, M.,Jin, Y. and Song, Q. (2019) Coloriza-
tion for anime sketches with cycle-consistent adversarial
network. International Journal of Performability Engineer-
ing 15(3).
[27] Zhu, J.Y.,Park, T.,Isola, P. and Efros, A.A. (2017)
Unpaired image-to-image translation using cycle-
consistent adversarial networks. In Computer Vision
(ICCV), 2017 IEEE International Conference on.
11
Towards Artificial Intelligence Serving as an Inspiring Co-Creation Partner
EAI Endorsed Transactions on
Creative Technologies
01 2019 - 04 2019 | Volume 6 | Issue 19 | e3
