There’s no universal definition of creativity, just as there’s no universal definition of love, but we all know it when we see it. So can a computer know creativity? More importantly, can we teach a computer to be creative? No human is born with abilities like writing, drawing, or singing—they’re skills we develop by observing and acting upon data, which we call “learning.” Similarly, AI can observe and learn from data to gain creative skills like drawing, musical composition, and writing. Machine learning is the gold standard for AI, and involves quickly and efficiently training an AI model on far more data - whether that data is text, music, images, etc. - than a human could take in a lifetime.
Can a machine be creative?
To an AI, creativity isn’t exactly the power of imagination as much as it is the power of creation. Artificial intelligence uses past data to translate a certain event or environment, learning from it just enough to generate new things. In the world of visual art, AI is making an impact in many ways. It can alter existing art such as the case when it made the Mona Lisa a living portrait a la Harry Potter, create likenesses that appear to be real humans that can be found on the website ThisPersonDoesNotExist.com and even create original works of art. Another development that blurs the boundaries of what it means to be an artist is Ai-Da, the world’s first robot artist, who recently held her first solo exhibition. She is equipped with facial recognition technology and a robotic arm system that’s powered by artificial intelligence. Another tool that embraces human and machine collaboration, AIVA bills itself as a “creative assistant for creative people” and uses AI and deep learning algorithms to help compose music. A powerful way dance choreographers have been able to break out of their regular patterns is to use artificial intelligence as a collaborator. Creative expression involves generating new ideas, which has always been understood as an inherently human skillset. But a growing number of research projects are demonstrating that deep learning, an advanced machine-learning technique, can be used to model and simulate many aspects of human creativity. ”
Neural net is a certain type of AI that simulates creativity. Anton Oleinik, a sociology professor at Memorial University of Newfoundland published an article explaining why AI will never outperform humans in creativity. There are multiple reasons on why neural nets are terrible at being creative, the first being their inability to recognize when a pattern will change. Neural nets learn algorithms by absorbing a vast amount of information and identifying every pattern in that data. They come up short when it comes to matching one pattern to a different pattern and anticipating when the pattern will change direction. Another big reason Oleinik recognizes is the fact that neural nets lack social intelligence. “Innovations are often embedded in social connections and relationships,” Oleinik writes. They are unable to see beyond the specific data set and its purpose. “Creativity is hardly possible without one’s capacity to think metaphorically, to coordinate proactively and to make predictions that go beyond simple extrapolation,” Oleinik says. AI will never be conscious and therefore it will never have the innovative thinking and spontaneous thoughts that change the direction of art. Artificial intelligence mimics creativity to create art from previous work. It works off of other people’s data. It does not have the capability to read human emotions and create demanded art based on social norms.