CHAT GPT & GENERATIVE AI, AND WHAT THIS MEANS FOR DESIGN EDUCATORS. Or, Why is ChatGPT and generative AI okay for industry but not for education?
Cover image is from the film They Live, 1988.
Our very own Sam Collett, who works as a lecturer in Graphic Design at University of Worcester, was asked to take part in a cross university conference on AI for educators. Below is the presentation as well as references and notes.
What became apparent as part of making the presentation Sam and Agate realised how much Machine learning based AI they use on a daily basis. Moreover we as a web agency rely on AI to generate content in the same way that just months ago we would have used copywriters, photographers and artists.
The full title of the conference is THE CHALLENGES AND OPPORTUNITIES OF ARTIFICIAL INTELLIGENCE (AI) FOR CREATIVE EDUCATORS. Therefore the talk is split between showing how agencies such as Practically.io are using these new tools, and what this might mean for teachers of art subjects. We do not pretend to have all the answers, and similarly we sit on the shoulders of giants when it comes to AI knowledge. But what we do have is that practice based knowledge that brings some realism to the talk. Take a look at the slideshow below, for a sneak peek of the talk.
SOME AI TOOLS WE USE (As of June 2023)
Transcripts > IBM Watson / LALA / OpenAI / Firefly
Music > LM – Google generative music
Images > Cariyon / StarryAI / Midjourney / Remove.bg / Runway / Adobe Cloud & Photoshop
Voices > Siri / Revoicer – voice overs with emotion
Copywriting > ChatGPT / Bard
WHAT ARE THE DANGERS OF AI FOR ART EDUCATORS? ACCORDING TO BARD
Artificial intelligence (AI) is rapidly developing and is already being used in a variety of ways in education, including art education. While AI has the potential to be a powerful tool for art educators, it also poses some potential dangers.
One of the biggest dangers of AI for art educators is that it could lead to the deskilling of art teachers. AI-powered tools could be used to create art, grade student work, and provide feedback, all of which are currently tasks that art teachers perform. If AI can do these tasks more efficiently and effectively than humans, it could lead to art teachers being replaced by machines.
Another danger of AI for art educators is that it could lead to a narrowing of the curriculum. AI-powered tools are likely to be most effective at teaching students how to create art that conforms to existing standards and expectations. This could lead to a situation where students are only exposed to a narrow range of art forms and styles, and where their creativity is stifled.
Finally, AI could also lead to a loss of human connection in the art classroom. AI-powered tools can provide students with feedback and instruction, but they cannot provide the same level of emotional support and guidance that a human teacher can. This could lead to students feeling isolated and disconnected from their art teachers and from the art community as a whole.
It is important to note that these are just some of the potential dangers of AI for art educators. It is also important to remember that AI is still in its early stages of development, and it is possible that these dangers will be mitigated as AI technology continues to evolve. However, it is important for art educators to be aware of the potential dangers of AI so that they can take steps to mitigate them.
Here are some things that art educators can do to mitigate the dangers of AI:
By taking these steps, art educators can help to ensure that AI is used in a way that benefits students and does not harm the art education profession.
WHAT ARE THE DANGERS OF AI FOR ART EDUCATORS? ACCORDING TO CHAT GPT
As with any technological advancement, AI can present both opportunities and challenges for art educators. Here are some potential dangers to consider:
Replacing human art teachers: While AI can help automate some aspects of art education, it cannot fully replace the value of a human teacher. If schools rely too heavily on AI for art instruction, it could lead to a reduction in the number of art teachers or a decrease in the quality of art education.
Reinforcing biases: AI algorithms are only as unbiased as the data they are trained on. If the data used to train an AI system contains biases, these biases can be reinforced in the system’s output. For example, an AI system that is trained on predominantly male artists may be less likely to recommend female artists to students.
Limiting creativity and experimentation: AI systems can be very good at predicting what will be popular or successful based on existing trends. While this can be useful in some contexts, it may discourage experimentation and risk-taking in art. If students are only encouraged to create art that fits within certain parameters, they may miss out on the opportunity to explore their own unique styles and ideas.
Ethics of AI-generated art: As AI-generated art becomes more prevalent, questions arise about the ownership and authenticity of such art. Who owns the copyright to a piece of art generated by an AI system? Is it ethical to sell AI-generated art as if it were created by a human artist? These questions will need to be addressed as the use of AI in art continues to grow.
Lack of emotional connection: Art is a deeply emotional and personal form of expression. While AI systems can mimic human emotions to some extent, they cannot truly understand the emotional impact of art on human beings. If art educators rely too heavily on AI-generated feedback or analysis, they may miss out on the important emotional connections that can be formed between teachers, students, and their artwork.
Notes, Image Sources and references
Write a witty 500-blog post on why AI will not replace humans. Write in the style of an expert in artificial intelligence with 10+ years of experience. Explain using funny examples
Generate 5 facts about “AI will not replace humans”
Who owns the IP of AI-generated content? https://medium.com/the-generator/ai-and-the-law-what-you-need-to-know-bfebb51b7d53
Digital art, poems, and books generated using tools like DALL-E, Stable Diffusion, Midjourney, ChatGPT, or even the newly released GPT-4 will not be protected by copyright if they were created by humans using only a text description or prompt.