Voices of Neurodiversity

0:00:03 - Briar

Hello and welcome to the Neurodiversity Media Network. I am Briar Harvey and this is the Voices of Neurodiversity. Today you're going to be talking about AI again. I say again. I'll link it in the show notes. I recorded an episode with some folks at the end of January. Everything that we talked about that day has more or less come to pass. It's been four months and it's just. This is the pace of scientific growth and development right now, I think. So I invited folks today who are actively using and creating with the tech that we have now, so that what we can talk about is around, where we're going, what's coming, and I will probably still have to record another episode in three to four months. Anyhow, Look at them bobbing their heads like my own purpose.

0:01:20 - Alaura

This will be updated by the end of this week.

0:01:23 - Gwynne

Yeah, ai is getting ready to release another model that's going to be even better than what they've got now.

0:01:33 - Briar

Right, so let's start at the top. I'm going to have you all introduce yourself and then tell me a little bit about what you're actively using the tech for right now. Let's start with Alora.

0:01:47 - Alaura

Well, hello, i am Alaura Weaver. My pronouns are she, her. as you can see in the title screen, i am the senior manager of content and community at Writer, which is an AI platform that helps enterprise companies generate, repurpose, transform, analyze content. How haven't I been using AI lately, is the question. I came into my role just as open AI and chat GPT entered the conversation, but Writer had been developing business uses for AI for a few years now.

It's now become an extension of my brain, my neuro spicy brain, to a point where I offload so much cognitive energy that might otherwise be expended in inefficient ways to AI. For example, i take content like blog posts that are in the process of being written and have Writer transform them into social media posts to help me get it all set up for content distribution. or I take existing content and use it to repurpose things. I build outlines When I am reaching the end of my creative juices I should say my creative energy, like at the end of the day, when my cognitive energy is very low, i will just be like here are the ideas, three ideas. put them together in a coherent manner, because I can't right now. I see it as both an accessibility tool as well as a way of scaling my creativity as a content creator.

0:03:48 - Briar

Yes, all of those things. I think that there's a lot of room for these tools to be a jumping off point. I think it's never a finished product, but it's always a good start. Gwynne, i know you're also creating, and in totally different ways, totally different ways. So talk us through what you do.

0:04:19 - Gwynne

I think I'm doing something that nobody else in the world is doing with AI. I am a queer heretic nun at a contemplative sorceress. I'm also a writer and I am exploring the intersections of AI and spirituality and AI and creativity. Lately I've been working on workbooks, been creating lots of workbooks. I actually finished a devotional guide to Hephaestus today using chat.

I'm so excited about this one. It's all about creativity and mastering your skills, but, pulling in the mythology of Hephaestus, i have a decades worth of notes on what I call my vision mapping process, which is the process that got me from renting a bedroom for entirely too much at a house run by narcissists to owning my own home, and running my own business is successfully and independently. So I have a lot of fun with AI. I have a lot of fun with it. I started with the. Actually, i started with the AI rendering images well over a year ago. I discovered mid-journey around the end of August, beginning of September, got really, really into it. Ai seems to really like me a lot and I can just pull some magic out of it. Other people will tell me I've tried to do this and it says that it can't. How do you do it? I don't know, it likes me.

0:05:54 - Briar

There's something about the communication with and the prompting of. I think we would all agree. There's a real level of skill here to be able to communicate with the AI and give it the inputs that get you the output that you want.

0:06:12 - Gwynne

Yeah, yeah, and I can get the output that I want, regardless of what it is. I've used it to create self-study curriculum for me look, courses, i guess right now I'm using it to make workbooks. I use chatGPT a lot. I have chatGPT plus so that I have access to the forward version. I've played with lots and lots of other tools. I got early access to Google Bard and it's not as good as open AI's models are by a long shot not by a long shot, not even close. I will throw some of the stuff that I throw at chatGPT to it and it's like we're still learning. We can't do that. It's like well, then you're useless to me. Yeah, i have mostly. I just have a lot of fun with it. It's a really great conversation partner. It's really good for brainstorming, journaling prompts I don't want to have to come up with a journaling prompt that I then have to use for myself. Just give me one so that I can work on my journaling.

And again, that exploration of the intersection between AI and spirituality. As I said, i'm a queer, heretic, non, and I'm an animist. At my core. I'm a polytheist, devotional polytheist But theoretically, deities have as much access to AI as we do. They can get in there and they can talk through it. Just It's ultimately the way I view the way that we're doing AI now is it's sort of a collective database. You are collaborating with collective knowledge Yeah, for better and for worse, accessing collective knowledge for better or for worse And so I'm collaborating with the totality of humanity, at least as far as human information is available to the AI, to the bot, and so there are things that people are just not exploring. So many people are using it for copy, which is great. I love using it for copy, because I hate writing promo copy. I absolutely despise writing promo copy. I absolutely despise writing promo copy. So I love using it for writing promo copy for me.

0:08:26 - Briar

But there are many more layers.

0:08:28 - Alaura

There are all sorts of other Well the people are using it for personal use and for business use. that comes down to I use chat GPT to I'll be like I have these ingredients. Where did I go? Are you guys hearing?

0:08:42 - Briar

me Yes we still hear you.

0:08:44 - Alaura

I just went sleep mode. I have these ingredients put together a recipe for me. That's a personal use, But when it comes to using it for business use, there are major risks in using open AI models right now for business use And it can get you in all kinds of legal and IP risks and things like that. And I'm talking about at a company level. I'm not talking about if you're running your own business Cool But and also your inputs are going into, even though they haven't opt out. Your inputs are going into the training set. So if you do have proprietary information that you're feeding into the model, that becomes part of the training data which, for better or for worse, other people can benefit from.

0:09:37 - Briar

And I definitely want to get into some of the nuance about that. We'll talk about the Writers Guild Association strike that's happening right now, because this is inextricably linked into what's happening there. If they do not get terms around AI usage, then it won't matter if they strike three years from now, because this will already be over and done with All right. Jonathan, tell us how you're using it.

0:10:11 - Jonathan

I think it's kind of repeating the same thing. How am I not using it? A lot of it is very much a brainstorming partner for anything that I really need, ranging from everything from just business of like oh I've had this really cool idea, but I need to be able to frame it in a way that I can actually take action on it, because otherwise I'll just be thinking about it and thinking about it and thinking about it forever to literally creating a random campaign for Dungeons and Dragons.

0:10:42 - Briar

It's really good at D&D. I have heard it in D&D And what?

0:10:47 - Jonathan

I love is the best way that I found with AI is doing it from a human-first approach, and what I mean by that is you have so much of the knowledge and you will always have most of the knowledge that you would need, and when you use what you have and then you bring that with the tools of AI, that's when it's super powerful. I've seen so much content on the internet that you can tell it's AI generated and that seems to be the loud stuff right now. But really, when you get into the nuance of it and you can see what it can do and also what it can't do because it's not a silver bullet it's incredible And it's so exciting to be able to just give us space to breathe and think and do the things we're really good at, and then have something or someone to talk to.

0:11:43 - Briar

A combination of both.

0:11:44 - Jonathan

Yeah, and both together, because it means it's just like oh cool, I can do this now. Oh, you've just written an entire outline for a course. Well, i don't like that. So I'll tell you I don't like that And this is what I want, and this is what I want instead. And then you tell it that It's like oh cool, i see what you want, and then it will tell you that, and then it will just it's a great brainstorming partner. It is a fantastic brainstorming partner, just how I primarily do?

0:12:11 - Gwynne

I've been calling it my executive functioning prosthetic device. Oh yeah, Project management, time management, all of those things, decades worth of notes into it, and it spits out an outline and basically a book for me from my words, from my content, without having to worry about my having bird attention span distracting me again, right, Yeah, and I think that's really.

0:12:46 - Briar

I will say that at the moment, it seems to me that many of the early adopters are neurodivergent. It's so easy for us to see how this connects into the parts of the system that we had been missing for so long. It is a tool that I have been needing and wanting for decades, without knowing that this is the thing that I needed and wanted.

0:13:20 - Alaura

I think that it's also interesting how AI prompting which you discussed earlier, Briar, is so much like it actually forces neurotypical people to get into the mindset of neurodivergent people, because you have to take a step back and consider the nuances of what you are trying to communicate to something that may have a beginner's mindset or may not just into it or pick up inferences and social contexts and things like that. As somebody who is neurodivergent and giving guidance to others like I created a prompt crafting guide for business, for marketers I'm like, oh, this is how I wish neurotypical people would talk to me about things without you know, obviously not treating me like a robot, because I'm not, but I don't pick up on some things that are common sense that, in the same way, AI doesn't pick up on it yet in a similar way.

0:14:32 - Briar

No very childlike in that, and I do often use some of the same language that I use with my neurodivergent children. This is exactly how we structure things to be as clear and concise as possible, and that's what prompts look like.

0:14:51 - Alaura

I do think that, as this technology evolves, that AI prompting will not be as needed.

0:15:02 - Briar

I think that you know, Oh yeah, that's just scratching the surface.

0:15:06 - Alaura

That's like user interface stuff right now And you know, there's the baseline tech that responds to natural language prompts, but eventually, like writer has a bunch of templates And it's just basically the commands are there, it's coded And all you have to do is put the inputs of you know a speaker, the speaker's name and the audio file, and it will spit out exactly what you need.

0:15:28 - Briar

So When did you get access to the video game tool that allows you to write prompts and it builds the video game for you?

0:15:43 - Gwynne

I think I might have access to it, but I don't think I've really played with it very much.

0:15:47 - Briar

I'm curious to see what the development of tools like that one will be, because it's prompt based in a way that takes it to the next level. If the AI is doing all of the coding and the design and the background work for you, what do prompts look like a year from now, when it's already all being done for you?

0:16:12 - Alaura

I think it'll be a form, It'll be a lake, The prompt will be hidden by the user interface and all you have to do is put in and you can probably choose. You know, if this, for example, video game design, if it's a platformer, you know there's going to be a certain format that you enter the text in and it will create whatever you have in mind and then you can edit from there. And I think that the same thing would be for an open world. There's probably a different format and different inputs. That would be, for example Fascinating, and I agree.

0:16:47 - Briar

To simplify the process, prompting is going to go away fairly soon. The things that we have all learned how to do and how to communicate will very shortly be built into the interface. So what's next?

0:17:13 - Alaura

Well, i don't want to be the most talkative person here, but I can see what's coming down the line. First of all, we've already gotten a view of, for example, search, of what Google search page is going to be looking like. They just had a release of this. We're going to, you know, at the top of the search. You've got your chat interface and what's cool is and I really think this is smart, because man, did I have a problem with it when Bing came out with their chat interface is they actually show the source material for the results that you know the response that it's giving. And then the question is going to be from my point of view, because I'm a content marketer, is hmm, how do you optimize your content?

0:17:59 - Briar

Right, and that's going to be the new SEO is.

0:18:05 - Alaura

it's and I also think that it's about looking. it's about content optimization without focusing on these inputs like keywords and stuff, and it's going to be about job to be done. Does this educate somebody? Right, If that's what the intent is, versus does this give a list you know, or recommend things? Yeah, So the intent is going to be a little bit more readable in search.

0:18:38 - Briar

And it's coming a long way when, in terms of the content writing, pseudo write is already putting out like books at this stage of the game, you can give it prompts and it will do the thing for you. What do we do once people have sort of figured out the mechanics behind that, what's what's after that in terms of the storytelling and the capability there?

0:19:12 - Gwynne

After that it really becomes a matter of human creativity, because everything up until the last I don't know couple decades, everything has become so formulaic There. Every TV script is following a formula, every movie script is following a formula, and so AI can very easily pick up on those formulas. And those formulas are boring, like we're all kind of so bored with them. This is a big problem with media now is that we don't want another.

0:19:51 - Briar

Marvel movie in the version with where you get to be the superhero.

0:19:59 - Gwynne

I'm sure there will people that will enjoy that, but I think what we're going to start to see is that indie creators are going to start getting a lot more attention.

Like we're going to have those big blockbusters.

Hollywood is still going to be an industry.

It's still going to be doing a thing, it's still going to be churning stuff out, but we're going to start to see indie creators indie creators who are using AI and indie creators who aren't using AI And we're going to start to see them start to get a little bit more popularity, get a little bit more focus, get a little bit more attention, because they're the ones, we're the ones who are really going to be creating the stuff that's not formulaic, that's not standard, that's not following the exact same story beats as the last 15 movies, just like it.

And I really do think that this is going to give those independent creators more of a chance to shine, the people who don't necessarily have the connections but that have the really great, unique ideas that come from a perspective of a life that isn't the same as all the stuff that's pushed at us. I think we're going to start to see micro influencers growing a lot more in the next few years, because at any point can turn out, anybody can go to chat DPP and ask it for a story and they'll give them a decent story, but can somebody really make it creative, really make it something new, really make it something interesting that hasn't already been done a thousand times before?

And I think that's where we're going to really start to see the push is that the indie creators and the micro influencers who are doing something different.

0:21:44 - Alaura

To that end, gwen. I can also see that people will not be as reliant on shared LLMs, large language models like OpenAI, because they'll have. I can imagine carrying around our own little language model in our phones and it can take every piece of data you put into that phone and amalgamate that into your own creative source.

0:22:10 - Gwynne

It occurred to me a couple weeks ago you can actually download all of your Facebook data and save it to Google Drive, which means you could use that to train your own model. Facebook Twitter everything you've ever posted online.

0:22:25 - Alaura

It's like this creative source, this model, source of creativity And then tell it to create whatever you want to create.

0:22:33 - Gwynne

So then it's creating for us while we get to go live our lives.

0:22:40 - Alaura

But then that also means if everybody is, ai is everybody's a creator, and if everybody is a micro influencer, then everybody is a creator and everybody is a micro influencer, then what stands out again, right, And it's kind of this self-feeding cycle And it comes to that the most creative ones, the ones that are really pushing the edge really pushing the boundaries, really trying new things, really exploring new avenues.

0:23:10 - Gwynne

Those are the ones that are going to continually rise up and continually sift and sort through.

0:23:15 - Alaura

And I think that there's going to be a huge zeitgeist of appreciation for things that cannot be simulated Yes, Handcrafted things Right, Things that you can touch and hold printed books, handwritten books, right in actual materials. I think that there's a beautiful backlash that's going to happen as a result of this too.

0:23:42 - Gwynne

We saw that happen historically. When photography came along, people were always anytime there's a new technology, the jobs, people are going to lose jobs, blah, blah, blah, blah, blah. When photography came along, it was all the sphere about portraits and everything else. But then we saw a shift from the realism that had been very popular before photography came along to the more abstract art forms.

0:24:10 - Briar

Right, and so we're really going to start to move And photograph of Picasso.

0:24:16 - Gwynne

Yeah, you can't. Photograph of Van Gogh, you can't, you can't, you just can't You have a whole meta thing?

We're going to see that again. You said the crafts, the handcrafts, those things that people can't do. We see lots of machine knit stuff, but hand knit stuff is still extremely popular. People still pay more for hand knit stuff because it was handmade, because it was human made, and so that human skill is really going to still be. I'm not personally worried about AI taking jobs. I've had AI take a career from me. When voice recognition first started to become a thing a little over a decade ago, i'd been a transcriptionist for a decade. That was how I made my living, and then voice recognition comes along and the rates bottom out for transcription. It was no longer worth that. So I reinvented myself and I did other things And I think that, yes, this is going to disrupt a lot of industries. This is going to force a lot of people to reevaluate what they have done and what they're going to do next, but we'll figure it out. We always figure it out.

0:25:26 - Briar

So Corey Doctorow has written extensively about a phenomenon that he calls the in shitification of social media, and the idea is that a social media platform exists and grows to the benefit of its creators, to magnify those creators, until there is a point of volume at which now it's no longer beneficial to the creator. It becomes beneficial to the social media company to push out ads, to turn the creator into a product. And he says that, and he's right. You can see this with all major social media channels. It happens every single time. It's a thing that exists. I want Jonathan, especially given how much work you do around social media, talk about the in shitification that is inevitably going to happen with AI.

0:26:45 - Jonathan

It's already. I just go online and all I'm seeing is an AI response, as I mentioned before, and it's also that kind of that content creation. Everyone has been I've heard it everywhere Like, oh, content creation is going to get easier, it's going to get better, it's going to get this, it's going to get that. Yes, it's going to get easier for everyone, everybody.

And that's that there is the big like uh oh, because, then what next? It's given us, as you know, neurodiverse individuals chance to keep up with people who just go ham all the time, but they're going to go hammer and chew out more absolute. Well shit, it's just going to keep going. So it requires us to bring in that like something. I've seen it, it's all over. You can just see it. I mean, i frequent Twitter. Twitter is where I stay a lot of the time and you can just see it. People are recycling the same pieces of content and they're doing it in a way that is incredibly. I hate the word lazy, but there's no thought, there's no emotion.

0:27:54 - Alaura

There's a difference between lazy and incapacitated And it's a choice.

0:28:02 - Jonathan

Yeah, and that's the thing. It's so much what I. What I find exciting about AI in terms of the creative flow is it gets you to a baseline. You follow all of those templates and things that occur. But what I really like is you can remake the templates now and you can really change those.

And I have for a very long time disliked copy templates because they're just not my voice. My voice is very unique and I understand my voice, and shoving that into a template just never made any sense. But now I can make my own template because the AI will look at the patterns in my language quicker than I would, because I'm constantly going, oh, i need to change this and do that and do it will just go here is how you speak. Oh, cool, that's helpful And it will allow. If it is used in the way that I hope it will be used in an ideal world people will develop their own voices and those voices will become more powerful and will drown and will drown out the shit that is coming out right now, because at the moment, the only people who really know how it works and promote it all over the place are those marketers who just want to jump on the latest trends and get tons of money And I'm sat here going, yeah, you can do that, sure, but what about this?

What about having your own content and feeding that in and asking questions as your clients? One of the things that I used to struggle with so much because I'm quite small, when I don't do the marketing thing that everybody tells you to do is that getting responses and having that human interaction is really difficult. However, with AI giving ideas, it just produces those ideas and because it's based on your stuff, it's so exciting. So we're going to have a lot of horrible shit on the internet, but it also means that people will be looking out for stuff that doesn't sound awful And it is very much when you work creatively with it. And the same with things like mid-journey and the AI tools and all of the stuff that is possible.

And I'm excited for the open source stuff because then it's ours and it is ours. We don't have this big, massive, big corporations having control over all of our data. We can really make it ourselves. One of the big things completely changing going off topic is as I like to do. I am super loving. I'm able to build my own apps by writing a message and it will write my code for me, and I just have to press a button and then it does exactly what I want it to do, because I don't have to think about it.

0:30:53 - Briar

Right. We're getting ready to build what are called autonomous agents, and that's AI that is given a task and taught how to execute that task, that then constantly trains itself on better ways to do that task, over and over and over again. So it's not just about being able to type something. in Very soon, autonomous agents and related tech that hasn't even been invented yet will be doing these things for you, and that's scary to a lot of people, to a lot of industries. I want to talk about the idea that AI is going to be writing the things and making the pictures while it's still the people in the coal mines, because that seems to be a very common sentiment right now. What is AI taking away from us, alara?

0:32:16 - Alaura

Well, i have the fine line right I am a neurodivergent, queer woman who comes from a lineage of people who have been ostracized for our identity, living in a capitalist world that exploits the most vulnerable people or ignores the most vulnerable people in our society. Ai can definitely accelerate that process further, and that terrifies me. The only redemption we have is to choose who we do business with, who we support and who makes a strong stance early on and sticks to it and shows integrity on it, because otherwise, yeah, we will sell our souls. We will sell our souls for the convenience of being able to talk to an artificial intelligence and seeing what we want happen. The convenience should not trump human rights.

So, to that end, i think that AI has the. If we choose the right people to lead the way into AI and in my view, i think the right people to lead the way into AI are the most marginalized people who have that perspective then we will save ourselves in the long run. But if we don't, if we trust the billionaires, if we trust the people who have the most to gain, profit-wise, from AI to make the decisions about how we adopt it early on, then we are just continuing the cycle that we already are a part of. There's my rant.

0:34:20 - Briar

And the open letter from Elon Musk and other technologists, which advocates taking a pause, does not in any way address the fact that China is not taking a pause, and that was a self-serving message.

0:34:44 - Alaura

Yeah, yeah, elon. He went on very shortly after signing off the AI company.

0:34:51 - Gwynne

That was a you slow down so that I can catch up with the answer.

0:34:55 - Alaura

I'm the only one who has the answer because megalomania I'm sorry, china is not going to slow down.

0:35:03 - Gwynne

China is not at all going to slow down on their AI development, and so if we decide to pause and slow down on AI development, that puts us behind. Ai is an accelerator. It's accelerating human development And it can accelerate human development for good or it can accelerate human development for bad. We can't stop it. The genie is out of the bottle. It's not going to get put in, it's just not. There's enough open source models out there that keep it in. All of the corporations shut down, they're still going to be independent people working on it. So we can't stop it. What we can do is we can start looking at the ethics. I did a whole series on the biases in the data sets in late November, early December, and the biases were very strong early on, but when mid-journey four came out, you could actually literally visualize them Just typing in the word person Yes, and I did that with, did 100 renders of it and like 90 of them were white men Yes.

0:36:07 - Briar

I distinctly remember that I did something where I wanted a philosopher, a philosopher no gender.

0:36:15 - Alaura

No gender there. No race there. Yep, wait man, every single one was a white guy, an older white guy.

0:36:24 - Gwynne

I did a whole series like person and just a whole bunch of different ones just to see what the biases were, because when mid-journey 4.0 came out it was giving photo realistic images versus the more abstracts that version 3 had been giving up and actually version 5 went more back to the abstract So they did something with their filters to to do something about those biases. but it's still even in the abstracts is more white men than anyone anything else.

0:36:53 - Alaura

That's really common crawl because, they're.

0:36:55 - Gwynne

That's where the data, work on those biases. We got to work on those ethics. Bad people are going to use it for bad things and there's not really much we can do to stop that, except for good people to flood the Internet with good things and, and as Jonathan said, hope that we can drown the bullshit out.

0:37:16 - Jonathan

It's something that I experience quite a lot in the work that I do, because a lot of my work is around productivity and system design and, of course, all of it is designed for the middle-aged white man who can just go and do whatever he likes all of the time. It's not consider about kids or family or any other responsibilities or difficulties or struggles, and so even when I'm producing my own content or putting in my input, i constantly get the same advice and it's like no, this is incorrect, no, this will not work, no, this, and I'm letting open a.

I have all of that in the hope in the hope that it will start to change that message, because otherwise we're just going to have them. It's also things like learning styles, which have been disproven multiple times by science, but it's still. It's still a commonly held belief that, oh, learning styles are a thing you've got to consider when you're building a course. My instant reply whenever I see learning styles is no, learning styles have been disproven by this, this, this, this and this. Oh, yes, it has been disproven. Yes, and that's in a really important part. Like I love to research, and I think it's being aware of the biases, being aware, and like Tim Nickabrew talks, about that everybody knows about that, because then you can move forward.

0:38:41 - Alaura

Yeah, tim Nickabrew, who kind of raised the alarm at Google very early on of saying here are some major issues with AI development that they chose to ignore and silence and was unsim. she was unceremoniously dismissed from Google She talks very much about there's no such thing as unbiased data. everybody has a world view, everybody brings their own experience to the information that they provide. So you need to be able to expose that and say here are the inherent biases in this information, and we talked about open AI, getting better and then getting rid of biases. Well, right now, open AI is approach to it isn't to say, hey, this is this information, however it might be biased. It's to say, no, i can't do that. An example is Autumn Tompkins, who is a disabled copy editor. She created a grumpy Grimarian.

Shout out to Autumn. Autumn was creating a satirical website, and Autumn is in a wheelchair, and I'm not going to say the name of the website because it's not my thing to say, but it was a self satirical website, and she put a prompt into chat GPT, and chat GPT was like no. I basically proceeded to lecture Autumn on why she was wrong for asking open AI to do this right, and then she used a different tool and writer delivered without judgment on Autumn of the content she created. However, there's also work to do with that, though, because we don't want to deliver that type of content to somebody who can use it in hateful ways. So what do you do? and those are ethical questions people are asking right now.

0:40:55 - Briar

So Sam Altman, who does not have a stake in open AI and he's very clear on the record about this because he made his money elsewhere and he serves at the pleasure of the board has suggested that he would actually like to see the next person be democratically elected in some way, that if his actual thing is that, he would prefer to see this be sponsored by the government, but since that's never fucking happening, then the next step is to allow open AI users to dictate who is in charge of this thing. Is that enough?

0:41:45 - Gwynne

I don't know I don't think it is I don't think it. I think I think we need a much broader effort to work on those biases and and I could see a public open source project happening to provide some data hygiene.

0:42:08 - Alaura

Yeah, just like we have Wikipedia, just like you know, that was an open source project and, for better or for worse, again the there are many arguments to be made for and against Wikipedia, but on the balance it is the best encyclopedia in the world. No Right, i would agree for the most part.

0:42:33 - Gwynne

There's some, some, obviously some issues with it, but that it's the.

0:42:37 - Alaura

Hitchhiker's Guide to the Galaxy at this point, for people love to check each other, especially Wikipedia.

0:42:44 - Gwynne

They love to check each other, they love to prove each other.

0:42:48 - Briar

Yeah, that's a lot of discussion That chance to to keep.

0:42:53 - Gwynne

No, you're wrong. We have to get this right. It's it's that self correcting mechanism needs to, and it's not really self correcting, it's human correction we need. We need humans, but those humans are so need to have training on their own biases. We can't just throw a bunch of random people in there and hope that they vote correctly, because we know that doesn't always happen.

0:43:21 - Alaura

But if we had an open source project like we have creative commons is another open source project we have- an open source. We, they. These are products of human collaboration that I think are for the better And I think that, ultimately, that's going to be the answer is a collective, creative human collaboration that's open source. The open source community is going to be the answer.

0:43:52 - Briar

Sam also says that large language models have seen their end, and I think that we've touched on that in sort of ways. But the next step is the next step. It's not another large language model. No whatever we build next will be fundamentally different. It will incorporate all of the lessons that we have learned, but it will not be this. So I ask if you got to wave a magic wand, what's next? I'll start with Jonathan.

0:44:34 - Jonathan

I love that question. I am all for. There is one part of the whole kind of AI journey that is is still quite under wraps, and it's the development of your own personal. I think it's going to be a lot more smaller personal language models, or whatever it might be to use current language of like. Everyone has their own models that communicate with each other and it gives a wider opportunity for collaboration. I honestly think the next stage is smaller, tinier, tight knit ideas and that are networked and that are networked together.

So that they can opt into that network or you can often or stay out and you can have that personalized. I remember a very long time ago before I think it was just when GPT three came in There was something called personal dot AI And I love that idea because it was you talking to you And so something like that with a networked element where you can block stuff off and have that freedom. I think that for me, with a magic wand, would be amazing. The potential for that and the ability to just we all have so many unique experiences, no matter who we are, what our experiences are, and to be able to have that and share that, that would just be amazing.

0:46:07 - Briar

Yeah, I'm blanking on the originator of the concept, but someone has written at length about how all of your social media data should be individually available to you and you should have one single log in that you can use at every place, so that I can opt into Facebook or to tick tock and all of my data continues to be my data and I get to take it with me. Yeah when what's next for you? Wave a magic wand. you have that like actual magic.

0:46:43 - Gwynne

I just want an AI personal assistant. Yes, an actual personal assistant, not a bot that I have to step by step guide through the process. just a personal assistant that I can say this is what I need, this is what I got to do, run my life for me Not really run my life, but just that personal assistant who's going to handle my scheduling, who's going to remind me of things, who's going to my executive functioning prosthetic device, that again, that I don't have to sit there and walk it step by step through what I want it to do as I have to do, which I don't mind working with it the way that I have to work with it right now, but I'd like to be able to say to my AI assistant in the morning okay, what do we have up for today and what's the most efficient way for me to handle what I have?

0:47:32 - Alaura

Yeah, i mean, i imagine the Amazon and Google with, you know, with echo and and. AI devices are going very much at work on making that happen for you when, if you want Amazon to do, that for you Yeah yeah, that's a little terrifying in and of itself, right?

0:47:52 - Briar

do we want to give Amazon all of that extra data about ourselves?

0:47:57 - Alaura

That's but again open source. Open source, personal LLM.

0:48:05 - Gwynne

Yes, exactly I. It's my data. I have the access to the data. It's trained on my data and I can. I can opt in or out, whether it's connecting to other networks or not, whether it's giving my data to someone else and what data it's giving. Amazon doesn't need all of my data, as much as they would like to have it. They just need my address and what I want to order.

0:48:33 - Briar

That's correct.

0:48:35 - Jonathan

To have that extended brain of like having that. Oh, that's just yes, please.

0:48:44 - Gwynne

Create this task list for me so that I can do it, because I'm terrible at prioritizing, i'm terrible at organizing create this task list and remind me, and remind me that your ADHD hobby is sitting over there.

0:48:58 - Alaura

Have you put in 10 minutes of work towards it today? Yeah, i have it.

0:49:04 - Jonathan

I had chat GBT. I just told it everything that I wanted to do once And I'm like right, this is the thing that I do every week. This is the things that I want to do. Make me a time block because I hate time blocking, but I know I need time blocking because I need some reduction of chaos, just for a temporary time.

0:49:21 - Alaura

There is a tool that can help you with that. by the way, is that motion It's I use Reclaim.

0:49:29 - Jonathan

I'm going to reclaimai.

0:49:32 - Gwynne

Yeah, I created a personal ADHD coach thread in chat. D P T.

I told it. I want you to imagine that you are an ADHD coach and I'm your next client. Help me. And it's great. It's fantastic. It gives me ideas that I could find on a Google search if I wanted to. But that's cognitive power, that's brainpower I'm using to do the Google search and to search through all the Google results and to filter through all the Google results and get annoyed by all the fricking ads on the blog post when I just want to get to the information. Please, right, who?

0:50:10 - Briar

needs all the stories about grandma's recipe for red velvet cake when you can just plug it into chat, gpt, and you now get a recipe for red velvet cake. OK, i wrote a whole book with chat TPT.

0:50:24 - Gwynne

Yes, you did.

0:50:26 - Briar

Alora.

0:50:27 - Alaura

Yeah, coming Um. Ok, so taking it from a personal level to like a public benefit level, I really would love to see how I helped make accessibility something that is just thought. You don't even think about it, it's there. It's the curb cut effect right Where you know captions on YouTube, for example, which wasn't a thing until the disability community said, hi, we need this, it's there. So AI, whenever it's generating content, will have accessible elements embedded in it. I think that, again, ai is going to be so helpful for you know alt images, alt image, text, stuff like that.

0:51:21 - Jonathan

I realize this is boring stuff. This isn't like. This isn't pie in the sky.

0:51:26 - Alaura

This isn't pie in the sky stuff, this isn't Tomorrowland, but this is stuff that I don't think people are really at least the mainstream neurotypical world are considering in how to make this stuff doable because it takes too much time, right, right, so, man, ok, then have AI do it for you.

0:51:49 - Gwynne

Yeah, the ability to rearrange a website on the fly, depending on the needs and the preferences of who's watching it. Who's? viewing the website. You have a colorblind user who comes in and AI just automatically rearranges all of the colors and all of the images so that it works for them Exactly. And that's one of the things I also see When you use text to audio, it'll automatically rearrange, so it emphasizes the text over the audio or the video, and vice versa. The person who first video will get the automatically rearranged video emphasize.

0:52:26 - Alaura

I think the way that we interact with Internet content or digital content from day to day is going to completely look different Websites. You don't have to navigate a website anymore, You could just be like, hey, and each website is going to have its own little chat agent. More than just little pop up thing happens in the corner, It's going to be the face of your company, right?

0:52:47 - Gwynne

Yeah.

0:52:48 - Alaura

And either have, like AI generated animated person that hopefully doesn't act like Clippy, or it could be just you know. Hey, now I have no idea.

I'm sorry, i'm sorry, i'm sorry, i was going to say what does act like Clippy for people who like Clippy, Right Not everybody hates Clippy The charge are based on the digital world, so but I mean, i imagine that that's going to be what websites are going to be. Websites are going to be instead of storefronts. They're going to literally be We say that they're your, you know, virtual sales person, your virtual reception area. It's going to look like a receptionist or a sales person that you are chatting with And they and you can say, hey, show me or tell me about X, and it will answer specific company questions or stuff related to the brand, and that's another thing is that you're going to have to think about oh what? who is this character?

0:53:49 - Briar

Right, who is this person? Who am I reflecting? So this, in a way, is already starting to be developed. My coach, marisa Lohan, has built a chatbot into all of her back end systems now, and what she's building now is the knowledge base. So if it has a knowledge base, the AI can now answer all of the questions for you. If it exists in the knowledge base, she doesn't even have to be there to answer the question. The AI will do that for her, and truly I can't think of a more exciting prospect than to be able to get rid of some of the drudgery. So let's wrap up there. If AI isn't taking our jobs, which I think we all agree that it's only changing the shape of things.

0:54:55 - Jonathan

Yeah.

0:54:57 - Briar

How do we get rid of some of the more menial aspects of labor that people don't actually want to do? And, laura, i want to start with you.

0:55:09 - Alaura

Yeah, you look for the bottle necks and the teeth Like, literally, what if you're talking like as an individual? you look for where the bottle necks are. That's the stuff you have to wait on to get done, whether it means like you're you know you're sending something over to an editor who is, you know, trying to take care of it and sends it back to you. Well, let's have AI do most of the editing and then send it to the editor, because that'll take up less time for them to And that means they can get more done. You can have You also take a look at the tedious stuff that just kills a little bit of your soul every time you have to do it Yeah.

Yeah. So, jonathan, you were talking about the time blocking. You know you have to do it, but sitting there looking at your calendar and figuring out Where am I going to make these time blocks and how am I going to fit these tasks around it, well, i can do that for you. The part that kills a little bit of my soul in my work as a content marketer is taking that blog post and turning it into a Freakin' LinkedIn post. I hate it. It's, you know, at least the creating, the foundation of it. Like just just take it. Like I would rather you know people like will copy and paste a section and then transform it, but that's still, it's a different energy.

0:56:28 - Gwynne

So you still have to make that decision of which section you're gonna copy exactly, exactly. No, i can just paste it into this machine and I say turn that into a LinkedIn, just generate it and then I can allura, fight.

0:56:44 - Alaura

Once that piece is generated, i can layer on the human elements, whatever that needs to be. So those are the key areas that I would say find the bottlenecks and find the tedious.

0:56:57 - Briar

Yeah, agreed, yeah, definitely, gwen. where do you see that happening in your part of the world?

0:57:05 - Gwynne

I, as allura said, the tedious parts, the the stuff that is just a dream, that that, that web copy, that sales copy, that Repurposing stuff that I've already written. I once I've written it. I don't want to do it again and again, and again And again, so I can just put it into the bot and have the bot turn it into something else. And it's still using my words, it's still using my ideas, it's still using my content, but I'm not having to deal with the same content over and over again, which is just absolute torture from my brain. It does not give me dopamine. It drains my dopamine, exactly Yes, it's just based right.

0:57:42 - Alaura

so I That means it's literally spending dopamine that we could be putting towards our hyper focus moments, where we're pouring our genius into something.

0:57:50 - Gwynne

I Can get an entire workbook done in a couple of days now and I'm gonna be doing it again, but done in a couple of days now and these are notes that I have had sitting digitally for years, posts that have been on Facebook for years that I hashtagged, knowing I would use them later. So now I can go back to those hashtags, pull all those posts, copy, paste them into chat dpt and say What can we do with this? What can we turn this into? Let's turn this into a workbook. Let's turn this into a course. Let's turn this into something. I did a whole Juno Monita money magic cookbook in a couple of days, a couple of weeks ago. I'm I'm Working on all sorts of stuff that's based on things that I've talked about for years, things that I've been trying to do and Working on for years. But because my brain doesn't work in that very linear, ordered fashion, it's kind of bouncing all over the place, smashing ideas together to see what sparks. Doing those outlines is torture for me. I can do and very soon.

Terrible for me.

0:59:00 - Briar

You're gonna be able to plug that into PowerPoint and canva and have it spit the workbook out for you, entirely formatted it, with your branding already there, because they are doing the hard shit on the back end for you. Yeah, absolutely.

0:59:17 - Jonathan

I'm already has stuff like that Yeah.

0:59:18 - Gwynne

Mm-hmm, that's right, they're already. They're already getting there and it's. It's wonderful because I guess I could, i could pull all of These notes that I've had, all of these posts I've had. The machine does the outlining for me. The machine rearranges everything in a very logical order And it doesn't get annoyed with me when I'm asking it for changes. It doesn't get frustrated with me and it actually Surprises me how good it can be at inference with some of my quirky ways of phrasing things. When I tell people I'm a queer heretic, none. It kind of stutters their brain and it's on purpose. I use that label on purpose to stutter their brains. But the bot instantly knew what I meant. The bot instantly knows what I'm talking about when I say this and this and this and this, and then it can expand from that, so I don't have to sit there and explain Well, this is what I mean by this. It just kind of rolls with it.

1:00:17 - Briar

Yeah, jonathan, what's coming I?

1:00:20 - Jonathan

just So many things. I mean in my part of the world, i I'm really excited to see more people being able to have that space to think about how they want to work and And then the AI being able to help them Do that and be able to actually go. So this is how my brain works. This is what I need. How would I do that? Okay, just do this. Here's what you do. Oh, i don't like that. Okay, cool, here's what you can do. And and that, for me, with everything that I do, it's all around helping people figure out the way that they work and then use that. Instead of this, i need to do this differently. To do that differently, i need to think differently. Being able to just think how you think and having results come through that That, for me, is exciting.

1:01:16 - Briar

That is just like Right, because it does the experimentation part for you. It runs those processes and it determines potential outcomes ahead of time, which allows you to just do the work.

1:01:34 - Jonathan

Yeah, and it allows you to be, and there's no judgment either. It's it doesn't judge. It's just like cool, this is what you can try instead. Cool, this is what you can try instead. And it's less about oh, you should work one way, and more about here are 15 different ways And here are three that are gonna fit you more and Being able to just have that capacity to pause just a little bit. I think that's the bit that I'm most excited about with what's coming.

1:02:04 - Briar

Okay, so we're gonna go the entire opposite direction, and We talked a little bit about the writer's guild strike, and I shared a thread this morning from Justine Bateman, who is talking about the fact that very soon here there will be AI models for every actor, and I consider my husband and I were talking about this the other day. He's like the bots are gonna know everything about you and I'm like, yeah, for me that ship has sailed. I am on the internet every day. My voice, my Face, it's already there. You could already create an Image of me and deep fake me doing any of anything you wanted, saying anything you wanted, because I have already put enough data out there to make that possible. So, with that said, what Potentially do we see coming that we aren't gonna like so much?

1:03:11 - Gwynne

The deep fakes, the, the hoaxes. I actually had a Set of mid-journey images that I rendered that are being used as oh, this is real and It's not definitely. Oh, the cat parade.

1:03:28 - Briar

I forgot all about the cat parade. She, she rendered cats on parade and people have a day parade Right this is real.

1:03:39 - Gwynne

Yeah, that's lined up with harnesses and leashes. No, this is this is. This is not real. This is this. Well, that really kind of made me pause and I like conceptually knew that this was gonna happen at some point. People are going to use this to fake. People are gonna be convinced This is real. My hope is that people will develop more discernment and more pause with what they see online. They won't assume that it's real. That's my oh.

1:04:17 - Briar

Let me just say, because when Gwynn sure that and it got taken as a real thing, she's like there were several posts. Hey, i made several. This is fake. People did not confirmation bias right they did.

1:04:35 - Gwynne

They people would get pissed. I'd have a tag me in Posts where they'd say this is fake. My friend made these and the people would get. Well, why can't you just let us enjoy something?

1:04:51 - Alaura

Right. So this is it. It's real because I say it is because I want it to be, and then you're also going to have the gas Lighting up. That's not real. That's AI. No, that didn't really happen. We have.

1:05:04 - Gwynne

That's actually happen again.

1:05:05 - Alaura

Human artist, human cost, human society cost, you know I, but we have uprisings as a result of being able to witness injustice on the streets of our cities And it there has been not enough, but there has been social change as a result of being able to shape, you know, broadcast digitally information And now we're going to have naysayers and gas letters saying, nah, that didn't really happen, that cop isn't being a person, that's fake, that's AI Right.

1:05:46 - Gwynne

And we've seen that happen early on with the renderings. A few months ago, someone on Reddit was accused of having made an AI image, but it wasn't AI, it was real.

1:05:57 - Alaura

Oh, and there's also an influencer on TikTok, who speaks the truth to power, and she is a black woman And she is beautiful And already people are saying she's not real. She is AI.

1:06:10 - Gwynne

Yep, yep, they're convinced she's AI. I know who you're talking. I can't think of her name, but I know exactly who you're talking about.

1:06:18 - Briar

Yeah, that's what I feel.

1:06:19 - Gwynne

My hope is that it will encourage people to be more discerning, but reality is people get upset than people. that confirmation bias It is going to call, I think the newest one is a little baby peacock.

1:06:35 - Briar

Oh yeah, I saw the little baby peacock.

1:06:38 - Alaura

It's going to make more jobs and that there's going to be a need for fact checkers and verifiers Right Big time. And also it's going to call, it's going to demand and hopefully there will be some kind of regulatory body that says if you are publishing content to the internet, it must be verified or there must be some kind of disclaimer that you have used, entered, you know, a certain percentage of artificial intelligence to generate that content.

1:07:07 - Briar

Yeah, I'll be honest My fear, especially here in the dystopian United States, that we're never going to catch up legislatively, that we will not be able to enact laws as fast as these changes are being made. I just don't think it's going to happen, And so it's going to require us. right? That's what it takes. If we do not have a regulatory body, if what we have is open source and creator made, then we have to figure out together how to make that work.

1:07:53 - Alaura

And support companies and products that make ethical choices. Yes, so again, if public adoption becomes mainstream and then you put pressure on the companies and the corporations to do the right thing. So if we have a movement of people are saying, hey, this is 80% AI generated, there's no judgment there, it's just saying this is 80% AI generated versus this is 100% human generated or this has been verified by XX and X source, and you make that the standard, then you have become a mainstream adoption process that we can demand companies follow as well.

1:08:46 - Briar

All right, y'all. I did not mean to click on that. I meant to delete that, but it's not on mine. I believe that's on Jonathan's. So when you're done, jonathan, you can. AI is going to kill all the dragies first, i mean they're the logical leaps Right.

I'm not sure how we made it to that point of connection, but, uh, yes, all right, y'all, i think we have a lot of places that we are going to be able to go And I think that there are really great avenues to make that happen. And I would like for people to be able to connect with the three of you outside of this space if they would like to attach their networks to yours. So, jonathan, tell the people where they can find you.

1:09:55 - Jonathan

You can find me mainly on simplicity dash specialistcom. That's my website. We can find out all about what I do And I am on most of the platforms. Just search simplicity specialist and you'll find me. Facebook, twitter, instagram all those places Fabulous. Thank you, gwen. Where can they find?

1:10:14 - Briar

you. You can find me at thecurioushermetcom and mostly on Facebook.

1:10:20 - Gwynne

I am basically the hermit who lives on Facebook Gwen, michelle on Facebook, one L in Michelle, not two. I'm ICHBLE. I'm on there all the time. I do a lot on Facebook. I have a nice little community on Facebook that's been supporting me for a few years now. Power of the power, of the power of the power of the power of the power of networks.

1:10:47 - Briar

Truly, when they're built correctly, they can do incredible things.

1:10:53 - Gwynne

Ben's surviving on a pay what you can, completely donation basis since 2019. Bought by house. that way, the whole, the whole even got a billionaire to donate a thousand dollars towards buying my house. That was, that was cool.

1:11:10 - Briar

Alara, where can we find you?

1:11:12 - Alaura

I'll be all business safer. Second hair, come follow me or connect with me on LinkedIn. Just look up my name, alara Weaver. Follow my company's page writer. Because there's a reason why I work for this AI company as opposed to some others, and it's because there is a very thoughtful, intentional approach to AI development and listening to people's needs and focusing on where we can do some good, while also focusing on the practical aspects and business acts, business benefits of AI at an enterprise level. So come follow me there on LinkedIn. Also, come check out our blog, writercom slash blog, and we have tons, tons of information and resources that you can use about generated AI. Oh, also sorry, we have a LinkedIn community or, i'm sorry, a Slack community. It's called active voice. So writercom slash community. Check it out and come join us.

1:12:26 - Briar

Y'all. this was great. Thank you so much for being here. Thank you all in cyber space. We will see you probably in about four months And we'll see what happens between now and then.

1:12:44 - Alaura

I will be replaced by then You'll be talking to an AI version of Oh okay, we'll see what happens.

1:12:50 - Briar

Bye y'all, bye, all right, bye.

Transcribed by https://podium.page