Grey Goop and/(or) Generative Synesthesia




(This is a funny Gemini graphic--it actually looks like the Grey Goop is going to wipe out the Generative Synesthesia.  Just a few prompts to create this visual, and I did not ask it to put these two concepts in opposition.)



Very much down the "rabbit hole" this week, considering how AI impacts human creativity.  Last semester in another class, I had to interview someone working in the field.  He is the learning scientist at the edtech I used to work at.  When we got to the topic of AI, which he and I both acknowledged is currently a ripe area of both anxiety and innovation in Higher Education, he said something that still sticks with me:  "That we have to be careful that we use AI to create, for instance, new modes of transportation vs. building a better horse-drawn carriage."  Reading some of the articles this week about IP issues and copyright laws, and a few about creativity and what is and is not "human-created", I turned to none other than Gemini to ask the following:  "Is there a body of growing research around the issue of using AI to further creativity or are we just building old models with new technology?"  From that query thread, I learned about "Grey Goop" (supported by several articles in the query thread linked below) and "Generative Synesthesia" (Zhou & Lee, 2024).  Basically, from a creative standpoint, like in many other aspects, AI can be a massive productivity amplifier, but it also leads to a homogenization of expression, patterns of thought, and individualized creativity.  I think we all know about the AI overuse of the em dash (something I've been using forever) and patterns of three, for instance.  But I find it fascinating that it can also lead to an individual "group-think tunnel vision" in a way that a single person sitting in a room, reflecting on their experiences, looking for novel ways of expression might not.  Or, at least, their tunnel vision might be their own unique flavor of tunnel vision.  

This line of inquiry also connected with some other reading I'm doing in preparation for a new job.  I'm reading Daniel T. Willingham's Why Don't Students Like School, 2e.  I worked with Dan as his editor many, many years ago, and I was eager to read this book.  Back in the day, he and I published a Cognitive Psychology text together when I worked at Pearson.  In the first chapter with the same title as the book, he says, basically, that humans aren't great thinkers and there are a lot of disincentives to doing the hard, effortful, uncertain work of thinking.  But, if we are faced with a challenge that is novel, and doesn't seem too hard, so we can achieve success, we'll apply ourselves and dig in to do the work.  I think the fundamental premise of the book, not surprisingly, is that education today either bores students (most often) or "over-faces" them with too difficult tasks without proper support. The connection--apparently, even artists, writers, and inventors might actually take some shortcuts through AI due to the challenge of trying to think of something new or find a new approach or perspective to an age-old problem.  This can lead to our collective production of content being like "grey goop"--or many versions of vanilla ice cream.  That's a bit of a stretch, but it does make you worry more deeply about what we lose in the context of such a powerful technology.  No question, where this research leads suggests that we always need a human in the loop.  But will that always be the case?  And what do we lose in terms of our humanity along the way, if that comes to pass?    


https://share.google/aimode/3Yb3IgX98osHz4eAF

1.     Zhou, E., & Lee, D. (2024). Generative artificial intelligence, human creativity, and art.                     PNAS Nexus, 3(3), pgae052. https://doi.org/10.1093/pnasnexus/pgae052.

2.    Willingham, D. T. (2021). Why don't students like school? A cognitive scientist answers                         questions about how the mind works and what it means for the classroom (2nd ed.).                       Jossey-Bass.

Comments

Post a Comment

Popular posts from this blog

Image
Musings on the readings:  First, the concept of a network vs. a community.  At first, I initially thought they were largely interchangeable, but some key distinctions can be made after doing some basic research on this.  Communities are often more about caring and providing support.  They also likely involve more time to create and care for the connection within the community.  Often they are smaller.  Communities can link to networks, and people within a community can provide those connections.  Networks are more transactional in nature, and as a result, people can come and go, get what they need, make a contribution, and then disconnect.  A community suggests a longer-term presence.  Professionally, networks sound more appealing to me.  I'm happy to make contributions and gain knowledge from a network within my professional setting, but I'm less likely to want to join a "work community".  I'm involved in several communities person...
Read more

Week 1 Notebook

Image
General musings as I read:  Two readings so far, one from 2008, one from 2011.  Certainly, some foundational concepts here, but in some ways, this content seems quaint given the digital contexts in which most of us participate in our daily lives.  Even our phones tell us how much time we spend online and in communities in the Web 2.0 environment.  The Kietzman et al "Get Serious" article seems really out of date, even if foundational.  I just left an edtech where our marketing arm sucked up so many resources it seemed sometimes that the product development was a bit of an afterthought.  I don't think any executive now doesn't embed some deep strategy in their operations to engage with social media and marketing, to build market share, brand awareness, and get actionable customer feedback and analytics.  AI is definitely making this sort of community-building around a brand far more compelling with personalized interactions.  It is also interesting...
Read more

Crowdsourcing Latin American Politics timelines

As someone who used to publish course materials for Latin American Politics courses, I loved the article by Wilson (2018) where the class was assigned a long-term project over the semester to crowdsource timelines for Latin American countries.  One of the key course goals in that course was teaching about political development in general, and how it occurs specifically in different countries in that region.  This project reflected that key course goal in a really unique way.  I've done similar activities in this program where the class was asked to annotate a shared document, for instance, building out a complete conceptual map of a topic.  I didn't love the way we had to do that project, but as with the project in this article, it was a really interesting way to get people engaged with their learning.  What the project in this article it was how the researcher showed that the project itself led to far greater learning opportunities and outcomes for the students...
Read more