00:11
So, we had our detour, as usual. The whole last episode on...
Only one this time. What a great detour.
Only one. Yeah, yeah, yeah. But I think you had the interesting terminology brought up in the
episode, two episodes ago, where you mentioned collective intelligence being one of your
takeaways from your climate frisk. No, climate fresk.
See, I told you. I'm already doing this.
Fresk. I'm already doing it.
It's so hard. I did it as well, so it's... Yeah.
But yeah. What?
Yeah. Okay.
Yep. Anyway, fresk.
Collective intelligence and what you think about it. Like, what sort of other things do you think
in relation to that? Yeah. So, when I was reminded of that term,
as we were talking about climate fresk, and I sort of remembered what I experienced hearing it,
which was, oh, that's a great term for the approach I take to a couple of classes or
even my teaching method. It's kind of how do I promote and facilitate the people, right?
The other people that I am trying to help learn about a particular topic, how do I help them
sort of share, engage? People use things like cross-pollination of ideas, stuff like that.
But how do I help that arrive? And in my space, as definitely more of a facilitator,
I consider myself more than a lecturer, right? The goal is to get that
river of thought flowing from people. And it came to mind because I'm doing that specifically
in the higher level classes, and I only have one of those. And it's a course on writing and
large language models, right? So, AI, like GPT. And so, what I chose to do with that class,
especially because kind of like climate in this space, it's an extraordinarily complex field
in the technical sense because of how it has developed to this point. And I am not an expert
in the design of LLMs. And it is also a highly complicated field when it comes to
the ways in which it's being picked up, used, abused, integrated, poorly integrated
03:01
into different things. And on top of that, you've got what I think is a huge part of the last,
you know, well, all of human history, but at least the last, you know, decades or so,
of the question of like, why are we doing this? And is it harming other people?
So, there is a large ethics discussion in there. But me being not an expert in those fields,
but having enough of my sort of space of intelligence, right, being aware of these
things and having a curiosity for the space, and enough information that I think I can guide
students to explore it, I emphasize the class as an exploration into that space, which is
essentially a collective intelligence space, right? We've spent the class doing a lot of
activities that are essentially either reading through sort of a smattering of materials that
I've kind of curated in some way, and I've taken from other like sort of curated spaces and pointed
them in those directions. And I've let them ask each other questions, ask me questions,
share their own thoughts. We've created a whole bunch of sort of comparison and pros and
consideration type tables about different topics on these things. And I have seen a number of kind
of perspectives that I hadn't even picked up on, which I think is where that importance from
collective intelligence comes in. Like, I'm not entering it in the space of I know all this,
I'm entering it with, we know a lot of things and can bring our own sort of pieces into play.
But this brings us to the sort of LLM, and I think chat GPT, and Google BARD, and Cloud AI,
and Sora for anyway. So it brings us into that space that I think we can talk about now,
if we'd like. So you said there was a chat you might have had recently, I don't know if you'd
like to. Well, that too, but like listening to you describe about how you're trying to incorporate
this idea of collective intelligence in your classroom. Yeah, I think it's interesting. I
always sort of thought that the team that I prefer to work in, as far as my research,
because I only have research environment type work experience, that's all I can speak of.
I think it's really critical that the collective intelligence is taken advantage of. But also,
it doesn't work just by putting a bunch of people with different ideas in the same room.
06:04
You have to create a space where people feel safe and supported to share those ideas,
and taken seriously, but also not too seriously, right? I like the space where I'm allowed to
blurt out stupid things as I think. It's a good space, yeah.
And have somebody else in a team be like, Asami, that makes no sense. Or I think you're not making
sense, but you're onto something, right? I like that kind of environment where... And that can
come from people in the same level, or people who are my bosses, or who are my supervisors,
or it can come from younger students or younger co-workers. It doesn't really matter where it
comes from, but I think I, at least, think I operate better in this kind of environment where,
yes, I need my space to think in sort of a solitary confinement way.
Especially when your ideas don't have words yet.
I think it's important. I have a hard time sharing ideas at that level of germination.
But once it's germinated, I would like to have people to sanity check me.
And bounce around ideas. And maybe if you are super experienced researchers,
this is what you do at a conference with people in a larger field. I don't know if I am
there yet. Conference still scared the fuck out of me. So I don't know. I mean, I hope
that I will quickly get out of this phase of being scared of conference, and actually enjoy
having discussions instead of being worried about being grilled. Right? Why not? But at least in
your home lab, in your home environment, I do think there needs to be sufficient diversity
in idea space and thought space so that people have meaningful bouncing around,
because if it's an echo chamber, there's no point. But that has to be not a hostile environment where
you're worried about being grilled, made to feel stupid in that. And I guess I was lucky that in
09:09
my grad school, I felt like that environment was already there. And thanks to my 22-year-old
instinct, I sniffed it out from a number of labs that I rotated. They're like, okay,
this lab seems to have that kind of environment. People feel safe to share premature ideas and
feel supported. So yay to my instincts. But I think it's a different space in the lab now.
I think right now we're still trying to find like a common language to share our ideas and
projects, because we are all involved in quite vastly different projects. And sometimes the
effort to have to share all of the context required to have the discussion is too much
that you kind of give up on sharing that further. You're like, okay, I would like your input,
but also it's too much to explain now. So maybe later. I don't know where I'm going with this,
but I think how do you sort of facilitate, how do you create that environment or try to?
I really appreciate hearing your experience with it, because there is,
I think, a sense that develops over time for people, right? What space feels like you can
have that, right? So like you were able to identify. Yeah. And by the way, I want to preface
this by saying that by no means, this is like the best for every team. It's just my personal
preference. I'm sure there are lots of different team dynamics that works better for plenty of
different reasons. I like this kind of environment. Yeah. I mean, you're not going to hear me sort of
complain or disagree. My whole sort of personality maybe revolves in that space, or at least I try
to. And maybe different styles work better, but I think what works with this style is a space where
ideas flow a little bit more freely and are given some room to breathe. And that means that you get
ideas that either turn out very impressive, not even, I wouldn't label them good or bad in this
case, but like really sort of nuanced or creative or something. And the ones that might either not
12:04
have that or perhaps be kind of, I don't want to say dangerous, but like in that space of
discomfort, right, they arrive and they're allowed to sort of be acknowledged and then let go of,
right? So the space sort of hopefully propels forward these sort of positive, even in complicated
and difficult situations, like solutions, ideas. And it's hard to share failures, right? Yeah. Science is
not designed to share failure. It is sadly enough designed to share, yeah. And I think
maybe because we work on sort of very independently different project in my current team,
I only hear highlights of their progress, right? The only way I can gauge that they might be
struggling is when they're reporting on the same thing, like they're stuck somewhere and they're
like doing the same thing or they have no new results or no new data to chew on. That's when
I'm like, oh, maybe we can talk, just bounce ideas off. But like, other than that,
like very few opportunities to be like, so this didn't work and that didn't work and that didn't
work, right? Yeah. Yeah. Which is what I did every day in grad school. I'm like, okay.
Okay, boss, I tried. I struggled for two weeks. Right. And here are things I did and none of it
worked, right? Yeah. Please help me. What is happening here? Yes. Yeah. Yes. And that space
is so, so helpful for that sort of, I think it's helpful to link back a little to your
mention of like, this might have been outside the podcast, but the idea of like,
being able to create one's own questions, it comes from feeling that the space is open to your ideas
and your ideas can turn into questions, right? You need to be ready and comfortable and,
and supported in that space. Otherwise you have to have on top of that an extraordinarily firm
sense of self to do it without a supportive space. Yeah. Like a sense of confidence that I know this.
There is, there is also in that space very close to the undeserved, you know,
confidence that many people might have about things. So the line is very, very fine. Yeah.
And also like, if you, like specifically for research environment, like if you know everything,
why, why are you researching? Yeah. Like, yeah, you should like anything worth doing research on,
you should be doing it because you don't know it. Right. If you already know it,
someone else probably also know it. And then what's the point, right? You've already,
15:00
you've got the answer. Now the idea is why you have to ask the question, why am I looking into this?
Yeah. You have to come up with a new question or synthesize question. And, uh, like, I think,
I think science should keep people humble in that way. Like it's, it's by design, but I do see a few
cocky ass people. Yeah. Science should perhaps keep people humble. People come to it with
attitudes that we develop because of social systems and because of situations. And, and
you also... That's another topic for another day. Speaking of other topics, would you like me to
answer the, how I create this space in a separate podcast episode? Or... I don't know. Do you think
you can do it in like two, three minutes? We can do that. I could do it in two and three,
if you want. Yeah. Okay. We can always dive deeper if there's a desire. So yeah. Yeah. In,
in two or three minutes, the sort of short version, I would say, um, my first starting point for
generating any space that is meant to be open and investigative and sort of safe for that is that I
make that intention clear. So like from the beginning of the, the classes that I did here,
it was, this is the way that I am going to run the class. Like the class is not about me
delivering unto you all of the riches of the world. Uh, it is, it is about me giving you
things that perhaps you have not seen before, hopefully giving you a bit of a foundation to
work from. And from those foundations, I want to put all of us into a situation
where you'll need to think about how this is affecting you, how it is affecting other people,
how is it affecting the space that you're in, the type of writing that you're in,
because this is still focused a lot around writing and sort of communication.
Um, and I make that clear up front and I try then as the sort of class and the space progresses
to embody that, right? I don't always, I think there are, uh, maybe more technical ways,
less technical, more structured ways to be like, you know, every class you can say that this is
how we do, and this is how we're going to do it. We're going to keep iterating on that.
I prefer a more natural like embodiment of it. So when I engage with the students,
I'm engaging with them as other intelligent minds in the room. And I listen and I stop and try to
take in what it is that they're sharing with me. And I spend a moment with that. I try to reframe
it. We try to collect ideas in different spaces, whether it be whiteboard, Google docs. And I just
say, just put them in there because the, that sort of removes the judgment that comes from it.
18:02
And it allows us to freely associate ideas in the space. You know, if somebody says something and
it comes up, we take it in and we go, ah, what about this other thing? Right. And it, it sort
of develops it over time by iterating on that environment itself. Um, that's maybe the origin
of it at least. So, so like, step one, um, intention, tell them, this is what I'm doing.
Tell them, get ready to get uncomfortable. Yeah. And step two, kind of let that simmer in. Yeah.
Let that sink in. Yep. Just see how they respond. Yep. And then step three, which I think is pretty
crucial, even though you just like went over it very quickly, like step three, sort of like making
it visualized and like shareable format, like that kind of seeds, seedlings of ideas. Um,
what you say means something we are taking. Right. Like make it, make it clear that
all of these little stupid ideas are being heard and are shared and have some kind of value. Yeah.
Yeah. Um, and, and, and by making that shareable, like on Google, what would you say whiteboard?
It, uh, apparently whiteboard or jam is no longer going forward, but it's just Google docs or like
the whiteboard, like the physical whiteboard. Whether it's put in words or doodles or something
like there's, it's, it's put in a format where it's archivable, like it's recorded and people
can kind of look back at it and sort of iterate over, iterate over those things because the
whole class then acts as one giant iteration. By the end, their final project is very free and
available for whatever they want to do, but it is sourced from the continued iteration throughout
the class, right? It's them considering their ideas, getting new ideas, like passing those along.
It's, it's a constant process as it goes. So yeah. Yeah. Cool. Yeah. So that's it.
Thanks for wrapping that up. I think we're, we're good.
Let's stop here before we divert even more.
That's it for the show today. Thanks for listening and find us on x at
Ego de Science. That is E I G O D E S C I E N C. See you next time.
