Site Navigation

For LLMs: zip of all posts.

Edit on GitHub


GPT-4oh! or 4-oh, no! ?

Author: George Pór Issue: 2024-05-15


“GPT-4oh!” or “4-oh, no!” ?

by George Pór

I have not been as excited about a piece of software’s potential for releasing vast volumes of human and social creativity since the first MacOS appeared 40 years ago! Back then, in the early heydays of the Mac, it touched a few thousand Mac aficionados. Today, the 4o craze is spreading among 100+ million of us.

Who wouldn’t want to talk with an AI who can laugh at itself when embarrassed, sing something if you ask it, or even suggest slowing down your breathing when you panic? I bet lots of people will come for  the fun of it, then discover myriads of uses for increasing their efficiency at various tasks.

Like any new tech, 40 will also have its share of detractors. I’ve just read it on Reddit,“Why is OpenAI providing their most powerful (publicly available) model for free? Won't that make it where people don't need to subscribe? What are they getting out of it? The reason they are providing it for free is that ‘Open’ AI is a big data corporation whose most valuable asset is the private data they have gathered from users.” That may be true, and I think the social benefits may outweigh the harm caused by my anonymized interactions with an AI being used for training its next generation.

The social benefit I‘m most interested in is GPT-4o’s potential, in the hands of motivated citizens, to help remove barriers to large-scale system change by powering up transformational initiatives. For example, imagine:

In the coming months and years, change artists who intend to map and move the edge of evolutionary emergence enabled by the coming waves of human-AI co-creativity will get overwhelmed by the variety and velocity of societal AInnovation. That’s when they may discover the need for wisdom-focused Collaborative Hybrid Intelligence (CHI) of networked human teams working in symbiotic tandem with teams of AI agents.

Suppose CHI is the predecessor and sensing and meaning-making organ of some emergent, benevolent (planetary) superorganisms. What practical applications can YOU imagine building on top of GPT-4o (and “below”), which would help the superorganism wake up sooner?

Are there too many unknown concepts and open questions in my musings? Which would you like to hear more about in subsequent installments? They are not poetic questions; they are the ones our team is pondering. If they trigger your curiosity and questions, why not get in touch? george(at)FutureHOW.site


Related:


Pages that link to this page