A Personal Perspective into the Future for Computational Creativity - Computational Creativity Research: Towards Creative Machines - page 392

aspects of process 8 requiring more complex assessment methods for novel artefacts,
based on models of human cognitive abilities [ 19 ], present a greater challenge. They
are nevertheless very likely to see a significant surge in the future, as our knowledge
of human cognition improves. I suspect that the implementation of all these design
processes in working computational systems will go a long way towards improving
the perception of creativity of the systems and/or their resulting outputs. This is much
in line with some of the existing descriptive models for creative systems [ 3 , 21 ], that
capture in their theoretical formulation concepts such as framing [ 3 ]ormeta-level
creativity [ 21 ] which correspond, as argued above, to instantiations of these design
processes.
This should not be taken tomean that the productive process—5 and 6, for applying
the constructive procedure to the ingredients and the selection criteria to the result-
ing candidates—are likely to be abandoned. By their very nature, these productive
process have a significant potential for applicability. A combination of processes
5 and 6 is the most likely candidate for finding commercial applications, where
instances of the produced artefacts can be sold to the general public, fulfilling the
need for personalised artefacts at low cost. It is important to note that the development
of new constructive procedures and new evaluation functions would correspond to
research on these processes. As such research is specific to each type of artefact,
and, within those, to genres or particular aims, it is unlikely that the potential for
innovative and valuable research on these processes be ever exhausted.
19.3.2 Factors that Will Influence the Future
of Computational Creativity
A very close discipline to Computational Creativity is Artificial Intelligence. Both
are based on computational principles, both attempt to emulate higher abilities of
humans, both were met at the start with mixed feelings of significant scepticism and
high hopes. There are some important lessons to be drawn from the evolution of AI
over time. The evolution of a discipline seems to be constrained by the expectations
it raises, the dreams it can inspire, the fears it invokes, and the financial profit it can
generate. All these come together to determine the amount of funding and effort that
can be invested in it. Another important point is that different sets of people react to
these various aspects differently. Researchers tend to be fuelled by dreams. Funding
agencies look more closely at expectations and financial profit. The general public
tends to consider expectations, and can be very much affected by any possible fears.
In the case of Computational Creativity, researchers are motivated by dreams of
machines capable of autonomous creativity, and of understanding human creativity.
These dreams should not be allowed to cloud the significant potential for impact
arising from other possibilities. Machine-supported human creativity, with machines
playing the role of colaborators or co-creators, may be very profitable in the medium
term [ 14 ]. Machine-creativity, different and beyond human creativity may prove to
be a very profitable outcome of Computational Creativity in the future [ 7 ].