TIME TO READ: 2 MINUTES
Feed in junk data and you’ll get biased, junk information
I’ll admit to being an Adobe snob so experimenting with Canva for the first time I was holding my nose before diving in. However, I was pleasantly surprised and had a fair bit of fun playing with the generative AI capabilities it offers in integrated third-party apps.
Sorry Adobe, but I have found that I quite enjoyed its user-friendly layout and quick, efficient design capabilities. There’s less of a learning curve compared to traditional design software, however there are limitations if you need to create more in-depth, tailored design pieces. But I can see why it is so popular. It is well-suited for social media design with templates to quickly customize to your needs. It also has a better price point than Adobe and is more affordable, especially for freelancers and small businesses.
I decided to try creating a video using the generative Magic Media function. A warning box popped up to tell me that this was “experimental new technology and scenes with people or animals may not look right”. So I settled on robots.
I’ve been reading Joy Buolamwini ‘s book UNMASKING AI. A prominent authority on artificial intelligence and a researcher at the MIT Media Lab, she founded the Algorithmic Justice League to create a world with more “equitable and accountable technology.” Her MIT thesis uncovered large racial and gender bias in AI services from companies like Microsoft, IBM, and Amazon. She demonstrated this in her TED Featured Talk on biases in algorithms by using a white mask with facial recognition software. With the mask, she is recognised, without, shockingly, her black skin is invisible to the algorithm. This highlights stark biases in the code the model had been trained on and evidence of discrimination and exclusion in tech products. An issue she calls “the coded gaze”.
As well as being a trailblazer in the field of Ethical AI, she is also a poet of code, combining art and research to highlight the social implications of AI and reduce its harms. Inspired, I tried my hand at AI poetry. I combined a short verse with the generative AI from Canva and created my own cautionary tale of generative AI and the risks of “junk” data.
And I used video footage of robots. Just in case there were any problems generating humans or animals.
©Jennifer Martin 2024
Jennifer Martin Design 