Artificial Intelligence (AI) is becoming increasingly prevalent in our lives, with an incredibly diverse skill set AI has become a driving force in both science and art but its effects on the communities differ vastly.
Illustration by The Geostrata
For ages, the debate about AI has been dominated by topics of its potential dangers: Algorithm bias and discrimination, the overtaking of human jobs and in the most exaggerated of scenarios the destruction of humanity, as AI is a reality and not just something that exists in a sci-fi movie.
More and more people are focusing on the positives implications that AI could have especially in areas such as medicine, climate science and green technology.
ADVANCEMENTS IN SCIENCE MADE BY THE USE OF AI AND ITS ETHICS
In a survey, the percentage of scientists worldwide reporting extensive use of AI in their research increased from 12% in 2020 to 16% in 2021. AI is advancing science in a range of ways by identifying major trends in large datasets and predicting the outcome.
AI is already advancing the research on complex neurodegenerative disorders like Alzheimer’s and Parkinson’s disease. In the future AI will likely develop beyond tasks like identifying patterns and designing experiments. If a decision made by AI has a negative impact, then there is no one to be accountable for it.
The most common and effective application of AI is in data analysis and processing, that is, AI will speed the analysis, quantification and visualisation of massive, complicated data sets. AI can quickly and accurately identify patterns or correlations that may not be derived from hypothesis-based research.
AI has advanced development rapidly when used as a coding or programming aid, it has already accelerated research in fields like natural language processing.
EFFECTS OF AI ON ART
AI’s implications for art are somewhat similar to that of technology and science but they can also differ greatly. Since the dawn of humanity, art has been a way of expression as well as a way of documentation.
Human creativity has been the driving force behind art throughout all of history, AI is shattering that paradigm. AI art allows anyone to create pieces or even collections of art within the fraction of the time that it would take by any other method.
When paired up with a human collaborator this technology can be used to create never seen before art but by itself. AI cannot create art of its own, it is always going to be a derivative of another piece. While human artists would use other art as inspiration, AI uses it as a sample.
TYPES OF AI MODELS
There are many different types of AI models which are used to generate art, including the following:
General Adversarial Network (GAN): One of the most common methods of creating AI art is General Adversarial Network (GAN). GANs consists of two networks - a generator and a discriminator. These work together to provide an image which is similar to the reference image. Over time, the generator trains itself to produce images that are increasingly similar to the reference image.
Image Style Algorithms: Image style algorithms are another way of producing AI driven art. This algorithm applies different image styles to each image, creating a unique visual effect for each one. This is done by combining two images- one for style and one for content- into one single image.
Computer Aided Drawing Tools: Computer-aided drawing tools have also been widely used in creating AI-driven art. These tools allow artists to draw and paint digitally, with the help of AI. AI gives suggestions for colours, textures, and styles. Here, AI is used to analyse the work of artists and give suggestions and guide the image towards the desired result.
Art Chatbots: Chatbots are computer programs used to interact with users in natural language. They can be programmed to generate text, images or sound. Art chatbots allow people to create images by providing information and guidance to the chatbot which, in turn, generates a response based on the given prompt.
ETHICS OF AI ON ART
The ethics of AI art are generally considered murky, a major concern being intellectual property theft, some large GAN-based AI art tools have been training data without obtaining the full copyright access resulting in multiple lawsuits.
For example, Getty Images sued Stability AI, a leading artificial intelligence company, alleging that their copyrighted content was used in the training of the AI.
This creates a dilemma for authorship because of the programmes that AI uses to create images, there is no sure-fire way of determining who created the piece: the AI, the artist whose work was used by the programme to create the image or the person that wrote the prompt to generate the work.
Humanity will forever grow alongside technology. Whenever there is a huge technological development it nullifies the need for jobs in industries that would now be considered outdated, but like any other technology, new different jobs emerge alongside it.
The potential of AI is undeniable, one of the most interesting things it has to offer is the automatization of menial tasks giving people the time to focus on tasks that require creativity and innovation, whether it be a scientific discovery or graphics for a video game. AI allows for the collaboration between humans and machines.
BY HRIDAYA
TEAM GEOSTRATA
While AI is indeed very useful, it still is, a double-edged sword.