Month: September 2023

Wk 6 – Automatic for the People

 

Week 6 Sketch Content

o Independently identify an interesting media work made with generative AI. Why do you think it is interesting? What makes it effective?

This work at the RMIT Design Hub Exhibition:

“12. Kate Geck Impossible Evolutions 2023 Digitally produced woven textiles Impossible Evolutions is a collection of textiles created in collaboration with machine learning models, offering speculative vantage points into the futures we share with critters and code. The models have been trained on datasets of endangered Australian butterflies and flowers to imagine new iterations. These ‘impossible’ machine-imagined evolutionary forms gesture to the futures we face as biodiversity decreases: there will be no further generations of these species. The woven textiles show the creatures as part of living systems: pollinators that enable new generations of plants that in turn provide food for new generations of insects. Revealing the interconnected relations of living systems as an important tool in advocating for planetary thinking. We put time and resources into developing machine-learning systems in the hope that they will improve many aspects of our lives and society. Impossible Evolutions explores how we might harness these systems to activate planetary imagination, advocating for hope and action towards a planetary commons.” (2023)

Wow. This work was amazing to view and explore. As it is described above, Kate Geck has created this exceptional work which is collaborated with AI and presented in textile form. She has used AI to generate images of flowers and insects that are extinct and/or not real, then she has displayed these images on a woven rug. Her work has shown me how AI can be beautiful and has the power to morph ideas and test boundaries.

Kate Geck Instagram Page

o With your group’s media sketch this week, what do you think worked well and what didn’t work so well?

  • Worked well =
  1. We had a solid story line that helped us piece together our video work.
  2. designating roles to each person also allowed us to expand on our work and make a sketch that is truely collaborative
  • Didn’t work well:
  1. Our videos do not run smoothly and would be very hard to understand with the narration, the characters and setting is not consistent.

o For future media works of a similar kind, what do you think can be done to improve the work?

  1. With more time and possibly with the paid subscription (for Runway), we could experiment further to create consistent characters and settings. However, I think that this may be more complicated than I think it is. ChatGPT is able to regenerate text with slight changes of your choice. I used ‘this person does not exist’ and each time I clicked the regenerate button, with the same prompt being used, a total new image would appear. In order to keep consistency, we may need to include an image as well as text in the Runway prompting tool, this way, the AI is more likely to generate videos with similar colours, brightness, figures ect

o What are your thoughts on this week’s reading? What is one key principle/thoughts/quote that really stood out for you — why?

Weizenbaum famously warned about the ELIZA effect where humans demonstrate awillingness even a desire to believe in computers, in his case as compassionate listeners.Weizenbaums warning about humanitys desire to be deceived is a central premise andanxiety about artificial intelligence.’

  1. AI has been experimented with since the 1960s… that is such a shock! There has been a lot of further development since then but I would have thought that AI began in the early 2000s.
  2. The anxiety in regards to AI started as soon as it began. It seems that people in general feel nervous at the thought of progressed technology and change.

 

Rewrite of this blog using ChatGPT:

prompt = Can you rewrite this as a five year old who is just learning to write

“Can you find something cool made by a smart computer? Why do you like it? What makes it good? I saw something cool at a special show:

It’s a bunch of pretty pictures and rugs made with the help of a computer. The computer learned about special bugs and flowers that are in trouble, and then it made up new ones! Even though these new bugs and flowers aren’t real, they look pretty on the rugs. This makes us think about how important it is to take care of nature. The computer also helps us imagine things for the future.

I thought it was super cool! It showed me that computers can make beautiful things and give us new ideas.

Now, let’s talk about our own work:

What went well = We had a good story that helped us make our video. We all had different jobs, which made our work better. What didn’t go well: Our videos didn’t play smoothly, and it would be hard to understand them with the talking. Also, the characters and places in our videos were not the same all the time.

To make our next computer project better, we might need more time and a special tool. This tool could help us make sure our characters and places look the same every time. But it might be a bit tricky!

Now, let’s talk about the stuff we read:

A long time ago, smart people started playing with computers, even before I was born! They were worried that sometimes people believe computers too much, like when they talk to them and think the computers care about them. This made some people scared of what computers can do.

It’s funny how computers have been around for so long, but some people still get worried about them. Change and new things can make people nervous sometimes.”