iSophist | Machine Rhetorics & Collaborative AI
banner
isophist.bsky.social
iSophist | Machine Rhetorics & Collaborative AI
@isophist.bsky.social
Enhancing Communication through Human-AI Collaboration

--> Join my newsletter: http://www.isophist.com
--> Follow my starter pack for Information & Content Development: https://go.bsky.app/LunWJCx
At first, I was excited about the new task feature in ChatGPT... then I realized they're making the same mistakes nearly every other task manager is making.
January 16, 2025 at 10:16 PM
Looks like all those students using AI to write academic essays might be saving the environment.

Recent research is challenging our assumptions about AI's carbon footprint. When it comes to writing, artificial intelligence might have a smaller CO2 footprint.
December 17, 2024 at 2:13 PM
With the help of AI and my love for coffee, I discovered one of the best analogies for understanding temperature in AI writing systems. ⬇️
December 13, 2024 at 2:50 PM
Today I tested a hypothesis that temperature settings with AI could be associated with different genres or styles.

It is false.

My experiments revealed something far more interesting. ⬇️
December 12, 2024 at 4:19 PM
A funny thing happened that perfectly illustrates why I'm cautious about using AI for research...

I recently discovered that Perplexity AI listed me and my colleague as "top 5 AI voices" alongside industry giants (after Ethan Mollick).

Flattering? Sure.

Accurate? Not quite.
December 11, 2024 at 2:07 PM
Meh ... Those are my thoughts about NotebookLM after experimenting with structured notes.

Check out the 🔗 to the whole video in the comments.
November 25, 2024 at 7:24 PM
What can a 4th-century saint teach us about building an AI knowledge base?

It isn’t about dumping documents and expecting magic. It’s a gradual process of curating, structuring, and connecting information bit by bit.

Like any craft, success comes from steady, intentional effort -- and patience.
November 20, 2024 at 2:17 PM
Perhaps the reason there is so much backlash against the use of AI in the writing process is because it threatens our (false) perception that we are all whole.

Excerpt From
Disability Rhetoric
Jay Timothy Dolmage
November 11, 2024 at 5:02 PM
Transformer Technology in AI isn't just about algorithms and data. It's about understanding context, crafting effective prompts, and knowing your audience - much like rhetoric.

By viewing AI through a rhetorical lens, we can unlock a deeper understanding of Transformer Tech
March 13, 2024 at 1:00 AM
This year I'm replacing the traditional multimodal assignment (podcasts, videos, presentations, etc.) with chatbot building. Here is my first crack at the assignment sheet. I'd love to hear some feedback!

🔗 in the comments.

#EduSky
November 27, 2023 at 9:40 PM
I thought I was going to do more prompt-building with students this semester ... but actually building them chatbots to help their writing was the biggest payoff.

So I was totally ready to write this when GPTs came out ...
I'd love to hear if anyone has things to add!

#AI #academicsky #edusky
November 15, 2023 at 2:42 PM
Do you want your GPT to start with the same greeting or sign off?

Use XML. Just be sure to tell it not to include the XML.

For my Medium Article Booster bot, I used the GPT builder, then added a structured frame around its instructions to keep it focused.
November 15, 2023 at 1:07 AM
Here is my No. 1 tip for a structured approach to GPTs.

⭐ Make one GPT per task.

The point of GPTs and customized chatbots is to narrow the use case of ChatGPT technology.

Know exactly what you want from the GPT ... then give it structured instructions for that task.

GPTs are tools, not people.
November 11, 2023 at 8:53 PM
I spent two weeks exploring with my students how to use AI to drive ethos-driven storytelling ... might as well make it a newsletter post.

#Edusky 🍎 #AI
November 6, 2023 at 2:15 PM
Blown away by Gamma's inline AI tool - a true cyborg sidekick!

I didn't want to write out a full definition, so I:

1️⃣ Added a slide
2️⃣ Titled "What is a Prompt Operations plan?"
3️⃣ Inserted info box
4️⃣ Hit ++++ button

Voila! It crafted a unique definition that meshed with my presentation.
November 5, 2023 at 8:18 PM
My kids are now creating their own personalized coloring pages with DALL-E 3.

It's a whole new creator economy. 😀
November 5, 2023 at 2:13 PM
Many noted the typo in this image. I'm impressed AI got that much right!

I used AI to transform a conference banner for LinkedIn.

1. Uploaded banner to ChatGPT.
2. Got a detailed description for an artist.
3. Gave it to Dall-E inside ChatGPT Plus.

Voilà! A new image with text was born.

#AI
November 2, 2023 at 12:04 PM
Ethos-driven storytelling in the digital age:

1️⃣ Create a persona of your reader; even better, interview them on Zoom!

2️⃣ Identify shared values.

3️⃣ Craft a story around that value.

AI can assist, but true connection goes beyond just chatbots.

Thousands share your story's resonance!
October 31, 2023 at 2:33 PM
I'm nearly done with my video course for @firehead.bsky.social, but in the meantime you can join me at the Best Practices Virtual Conference November 6-8, 2023. I'm leading a hands-on workshop, Prompt Operations for Content Developers.

For additional information, visit the link in the comments.
October 29, 2023 at 12:16 PM
I get weirdly excited when someone in our AI Governance committee starts drawing an object-oriented knowledge graph.

The best way to understand institutional AI governance is to map out the people, places, and things.

AI governance isn't a thing ... it is a network of relationships.
October 27, 2023 at 11:54 AM