Latest / 6 Ps in AI Pods (AI6P) / 🗣️ AISW #127: Santa Meyer-Nandi, Germany & France-based environmental lawyer

Transcript

🗣️ AISW #127: Santa Meyer-Nandi, Germany & France-based environmental lawyer · 6 Ps in AI Pods (AI6P)

• 0:00
  SPEAKER_01: Hi everyone.
• 0:00
  SPEAKER_01: This is Karen Smiley.
• 0:02
  SPEAKER_01: Welcome to 6P's in AI Pods.
• 0:07
  SPEAKER_01: This post is part of our AI 6P interview series on AI software and wetware.
• 0:13
  SPEAKER_01: Our guests share their experiences with using AI and how they feel
• 0:17
  SPEAKER_00: about AI using their data and content.
• 0:21
  SPEAKER_01: I'm delighted to welcome Shanta Hernandi from Germany and France as my guest today
• 0:26
  SPEAKER_01: on AI software and wetware.
• 0:28
  SPEAKER_01: Shanta, thank you so much for joining me on this interview.
• 0:31
  SPEAKER_01: Please tell us about yourself, who you are and what you do.
• 0:34
  SPEAKER_00: Thank you so much for having me here, Karen.
• 0:37
  SPEAKER_00: This time as your guest, last time we had a big pleasure of interviewing you.
• 0:43
  SPEAKER_00: So I'm an international environmental lawyer by background and I'm specialized
• 0:49
  SPEAKER_00: in international and EU environmental law.
• 0:52
  SPEAKER_00: But over the years,
• 0:53
  SPEAKER_00: my work has evolved into helping organizations navigate complex transformations with
• 0:59
  SPEAKER_00: a focus on sustainability.
• 1:01
  SPEAKER_00: So I co-founded Finding Sustainia with Anna,
• 1:04
  SPEAKER_00: who has also been on your podcast, and we've been working together since 2000.
• 1:08
  SPEAKER_00: With AI coming, we were at first really excited about the possibilities it could bring.
• 1:15
  SPEAKER_00: So we adopted AI in our own practice pretty early
• 1:19
  SPEAKER_00: and also started giving AI workshops for sustainability professionals.
• 1:25
  SPEAKER_01: So as an environmental lawyer, what first attracted you to AI?
• 1:31
  SPEAKER_00: As a person, I was attracted by the facility of all the potential.
• 1:36
  SPEAKER_00: But then with time,
• 1:38
  SPEAKER_00: the first little scandals came up and then I put on my environmental lawyer hat
• 1:44
  SPEAKER_00: and looked at first, of course, the environmental impact as such.
• 1:48
  SPEAKER_00: And you have also written a wonderful book about that.
• 1:52
  SPEAKER_00: But then I was also looking at the second and third order effects because this is
• 1:57
  SPEAKER_00: where it gets interesting.
• 1:59
  SPEAKER_00: So less about the question of what AI can do today,
• 2:03
  SPEAKER_00: but more about how it changes incentives,
• 2:07
  SPEAKER_00: leadership,
• 2:08
  SPEAKER_00: organizational culture,
• 2:10
  SPEAKER_00: and then on that third level, decision-making, society, democracy.
• 2:15
  SPEAKER_00: And this is where I started to feel we need to treat AI in the context of
• 2:21
  SPEAKER_00: the precautionary principle and the responsibility principle,
• 2:24
  SPEAKER_00: because we are already seeing certain effects on young people, on their brains.
• 2:29
  SPEAKER_00: There have been a few AI-led suicides.
• 2:32
  SPEAKER_00: So how do we handle that?
• 2:35
  SPEAKER_01: You've often talked a lot about systems thinking.
• 2:38
  SPEAKER_01: How does that shape your perspective on AI and these second and third order effects?
• 2:43
  SPEAKER_00: So of course, technology never exists in isolation.
• 2:47
  SPEAKER_00: Everything has its repercussions,
• 2:49
  SPEAKER_00: whether it's electricity or how cars were replacing horse carriages.
• 2:55
  SPEAKER_00: So the first order is of course productivity.
• 2:59
  SPEAKER_00: But then the second question is, what does it do to us?
• 3:03
  SPEAKER_00: How do people communicate with each other?
• 3:06
  SPEAKER_00: So if we look at meta AI,
• 3:08
  SPEAKER_00: it has actually connected us superficially, but has made us far more lonely.
• 3:15
  SPEAKER_00: Social time in the US,
• 3:16
  SPEAKER_00: there are some studies,
• 3:17
  SPEAKER_00: I imagine studies are pretty similar in Europe and other countries.
• 3:21
  SPEAKER_00: Social time has decreased by up to 80%.
• 3:26
  SPEAKER_00: And so all of these things have repercussions on us.
• 3:28
  SPEAKER_00: Then since Meta exists on a smartphone, teenage suicide rates and depression rates have gone up.
• 3:37
  SPEAKER_00: And all those things are concerning us systematically as a system.
• 3:41
  SPEAKER_00: And they're not just isolate little things,
• 3:44
  SPEAKER_00: but when those companies such as Meta or OpenAI are more powerful
• 3:50
  SPEAKER_00: than nation states,
• 3:51
  SPEAKER_00: we really need to look at trust, governance and democracy in a different light.
• 3:57
  SPEAKER_01: Tell our audience a little bit about your level of experience with AI
• 4:01
  SPEAKER_01: and machine learning and analytics if you used it professionally or personally
• 4:06
  SPEAKER_01: if you've ever studied the technologies.
• 4:10
  SPEAKER_00: Pretty much your classical user except for an informed one.
• 4:14
  SPEAKER_00: So I'm not an engineer.
• 4:15
  SPEAKER_00: I don't have your technical expertise.
• 4:18
  SPEAKER_00: I look at it more from that systems perspective,
• 4:21
  SPEAKER_00: but I also look at it as a tool for sustainability professionals,
• 4:26
  SPEAKER_00: for instance,
• 4:27
  SPEAKER_00: to deepen our work,
• 4:29
  SPEAKER_00: to democratize certain tools that have been inaccessible for a lot of NGOs because
• 4:34
  SPEAKER_00: of cost and lack of skills.
• 4:36
  SPEAKER_00: I'd say that's my relationship to AI.
• 4:39
  SPEAKER_01: What would be an example of a tool for sustainability professionals
• 4:44
  SPEAKER_01: that is now available that didn't used to be?
• 4:47
  SPEAKER_00: For instance,
• 4:47
  SPEAKER_00: there's a German startup that's called Regulai
• 4:52
  SPEAKER_00: and they are making different regulations easier.
• 4:56
  SPEAKER_00: It's basically work that would have taken weeks and months and running it through
• 5:02
  SPEAKER_00: a specialized AI tool is very helpful.
• 5:05
  SPEAKER_00: And that means that professionals then can allocate their time on far more important
• 5:12
  SPEAKER_00: and relevant things.
• 5:14
  SPEAKER_00: So they're doing much less of the legwork, but they can focus better.
• 5:18
  SPEAKER_00: So there's a lot of potential AI tools that can work as an early warning system.
• 5:22
  SPEAKER_00: Both of us looking at different tools that can really be helpful.
• 5:27
  SPEAKER_01: That's a good example.
• 5:27
  SPEAKER_01: So thank you for sharing that.
• 5:29
  SPEAKER_01: Has AI changed the way that you do your work with Finding Sustainia?
• 5:35
  SPEAKER_01: How has it changed your way of working?
• 5:37
  SPEAKER_00: There are things that used to take me hours and days.
• 5:40
  SPEAKER_00: I use it as a co-creator.
• 5:42
  SPEAKER_00: I ask it questions.
• 5:44
  SPEAKER_00: I use it as an editor as well.
• 5:46
  SPEAKER_00: It really helps to have this tool to improve my writing, but I'm very protective of my mind.
• 5:53
  SPEAKER_00: I don't know whether you've noticed it already.
• 5:55
  SPEAKER_00: A few days ago,
• 5:56
  SPEAKER_00: I wanted to write down a difficult English word and I was stumbling a little bit
• 6:01
  SPEAKER_00: because my first thought was, okay, AI can just do it.
• 6:05
  SPEAKER_00: These are the kinds of ways AI saves me and saves us a lot of time,
• 6:09
  SPEAKER_00: but we need to also make sure we protect our minds.
• 6:14
  SPEAKER_00: When you work with AI tools, are there specific tools that you prefer?
• 6:19
  SPEAKER_00: I like Claude.
• 6:20
  SPEAKER_00: I also like Notebook LM.
• 6:22
  SPEAKER_00: It just helps me reach reports sometimes in a bit more pleasant kind of way.
• 6:29
  SPEAKER_00: I sometimes really like to listen to certain material as a podcast.
• 6:34
  SPEAKER_00: So I really like those ways to deepen knowledge.
• 6:37
  SPEAKER_00: Not everyone is a visual learner.
• 6:40
  SPEAKER_00: Some people learn by hearing.
• 6:41
  SPEAKER_00: So I think there's a lot of potential collaborating with AI.
• 6:45
  SPEAKER_01: So when you think about the different tools that you use,
• 6:49
  SPEAKER_01: which aspects of it make you feel more optimistic and which ones are concerning
• 6:54
  SPEAKER_01: or make you more cautious?
• 6:57
  SPEAKER_00: I recently met a wonderful person who's creating an AI tool which actually lets
• 7:04
  SPEAKER_00: a ride by hand and it's a very beautiful embodied,
• 7:08
  SPEAKER_00: almost poetic way, which has a very rigid way of using the right kind of sources.
• 7:14
  SPEAKER_00: So that made me optimistic.
• 7:15
  SPEAKER_00: There's something very beautiful there and people can also write together.
• 7:20
  SPEAKER_00: So that's something that could enhance humanness.
• 7:25
  SPEAKER_00: And then of course the different tools I was telling you about
• 7:28
  SPEAKER_00: that help us do less leg work and that can go through reports,
• 7:34
  SPEAKER_00: that can assess risks.
• 7:36
  SPEAKER_00: So that makes me hopeful and probably also optimistic,
• 7:40
  SPEAKER_00: even though I'm also a bit careful using that word.
• 7:43
  SPEAKER_00: I don't know, do you feel optimistic with AI?
• 7:46
  SPEAKER_01: I think mixed is a good way to put it.
• 7:48
  SPEAKER_01: There was a recent article on Substack,
• 7:50
  SPEAKER_01: I don't know if you saw it,
• 7:51
  SPEAKER_01: but someone was talking about AI cheerleaders and AI doomers and then the people in
• 7:56
  SPEAKER_01: the middle who are more looking at it as AI redirection basically
• 8:01
  SPEAKER_01: that has this potential,
• 8:02
  SPEAKER_01: but there's also some harms that are happening now and we need
• 8:05
  SPEAKER_01: to not just ignore those harms.
• 8:08
  SPEAKER_01: We need to be doing something about them while we look at ways
• 8:11
  SPEAKER_01: that we can take best advantage of AI tools.
• 8:14
  SPEAKER_01: I'd put myself definitely in the redirection camp in the middle.
• 8:18
  SPEAKER_00: If I can imitate you, I'd say I'm exactly there too.
• 8:23
  SPEAKER_01: So sustainability is one of your personal areas.
• 8:26
  SPEAKER_01: One of the concerns that comes up, as you mentioned earlier, is with high sustainability.
• 8:31
  SPEAKER_01: On the one hand,
• 8:32
  SPEAKER_01: machine learning,
• 8:34
  SPEAKER_01: not so much the generative AI,
• 8:35
  SPEAKER_01: but machine learning tools have a lot of potential to improve energy efficiency and
• 8:41
  SPEAKER_01: to make better use of the resources that we have.
• 8:44
  SPEAKER_01: But on the other hand,
• 8:46
  SPEAKER_01: the way that a lot of the tools are being used now,
• 8:48
  SPEAKER_01: it's creating sustainability issues.
• 8:50
  SPEAKER_01: And so we have some trade-offs there.
• 8:53
  SPEAKER_00: I don't know whether you saw the data.
• 8:55
  SPEAKER_00: Most ChatGPT users actually just
• 8:57
  SPEAKER_01: use it as a friend.
• 9:00
  SPEAKER_00: I personally would like it if one email cost us one cent or one WhatsApp message.
• 9:05
  SPEAKER_00: Use is too cheap,
• 9:06
  SPEAKER_00: so it also invites in mindless use, which is then also costing us water and energy.
• 9:14
  SPEAKER_00: So because people are getting very lonely,
• 9:18
  SPEAKER_00: they're using it as a crutch, which then can make us even more lonely.
• 9:25
  SPEAKER_01: Yeah.
• 9:25
  SPEAKER_01: And to your point about thinking from the systems perspective,
• 9:28
  SPEAKER_01: there was already a loneliness problem before Jennifer A.
• 9:31
  SPEAKER_01: Atolds really broke onto the scene in a big way.
• 9:34
  SPEAKER_01: It may feel like a solution,
• 9:35
  SPEAKER_01: but in the bigger picture,
• 9:37
  SPEAKER_01: it maybe is not because as you said,
• 9:39
  SPEAKER_01: people end up feeling more isolated and more lonely as a result.
• 9:43
  SPEAKER_01: Not solving the problem is actually making it worse.
• 9:45
  SPEAKER_01: Those are some of the things that I'm imagining that you're looking at.
• 9:49
  SPEAKER_00: Yes, I find it very interesting.
• 9:51
  SPEAKER_00: AI has greatly contributed to the fact that we don't sit with vulnerability
• 9:55
  SPEAKER_00: and uncertainty that much anymore.
• 9:57
  SPEAKER_00: I forgot the numbers,
• 9:59
  SPEAKER_00: but an incredible amount of young people have their first romantic experiences with
• 10:04
  SPEAKER_00: an AI girlfriend, particularly with Meta-AI and other agents.
• 10:09
  SPEAKER_00: And what does that do?
• 10:12
  SPEAKER_00: It helps us to just use it as a crutch to not make ourselves vulnerable.
• 10:16
  SPEAKER_00: We have this available fictive persona that's just there and how can
• 10:23
  SPEAKER_00: a real person ever match up to that?
• 10:27
  SPEAKER_00: That would then again make us far more lonely because it's not something real,
• 10:31
  SPEAKER_00: we can't hold their hand
• 10:32
  SPEAKER_00: and we both know how fights can actually make us stronger even though in
• 10:37
  SPEAKER_00: that moment it feels icky.
• 10:40
  SPEAKER_01: Yeah, there's a lot of discussion in learning and in relationships that some amount
• 10:45
  SPEAKER_01: of friction is needed to build up the strength to develop those skills.
• 10:50
  SPEAKER_01: And without the friction,
• 10:52
  SPEAKER_01: the skills just don't get built and then they don't do well in interactions
• 10:56
  SPEAKER_01: with other people.
• 10:58
  SPEAKER_00: Yes.
• 10:59
  SPEAKER_00: And now even in dating apps,
• 11:00
  SPEAKER_00: there's that phenomenon of being chat fished,
• 11:04
  SPEAKER_00: which means that without falling into gender stereotypes,
• 11:08
  SPEAKER_00: you may be having the most thoughtful empathetic conversation with someone.
• 11:14
  SPEAKER_00: And then you meet the person and they don't even ask you one question.
• 11:18
  SPEAKER_00: These are all individual questions.
• 11:20
  SPEAKER_00: It might make you feel less vulnerable that you have that sparing partner.
• 11:24
  SPEAKER_00: It just helps you always sound polished.
• 11:26
  SPEAKER_00: But if you look at it at scale as a real question, what that does with us as a society.
• 11:33
  SPEAKER_00: And our frustration tolerance and all these soft skills we sometimes need to acquire
• 11:39
  SPEAKER_00: the hard way.
• 11:41
  SPEAKER_00: And I can also see it with my students.
• 11:42
  SPEAKER_00: I teach at Essex Business School in France and I was shocked this time
• 11:48
  SPEAKER_00: when I was teaching my students.
• 11:50
  SPEAKER_00: They were twitching because I'm quite strict on smartphones.
• 11:53
  SPEAKER_00: You could see they were trying to do everything with ChatGPT.
• 11:58
  SPEAKER_00: Then I had to give them an assignment.
• 12:00
  SPEAKER_00: They had to write it by hand.
• 12:01
  SPEAKER_00: But then what they did is they generated and then wrote it by hand.
• 12:05
  SPEAKER_00: And then I had to go through some better and less comprehensible handwriting.
• 12:11
  SPEAKER_00: I had a student who read her little script for a presentation in 30 seconds and
• 12:16
  SPEAKER_00: then started scrolling in front of all of us.
• 12:19
  SPEAKER_00: And there's this numbing effect.
• 12:22
  SPEAKER_00: There are some brain scans that have been made from children and from young people
• 12:27
  SPEAKER_00: and their brains are similar to those of 60 year olds or older ones.
• 12:31
  SPEAKER_00: And when you think of wetware,
• 12:33
  SPEAKER_00: I've learned that term from you, what is this kind of AI usage doing to our wetware?
• 12:39
  SPEAKER_01: Yeah,
• 12:40
  SPEAKER_01: I hear a lot from different teachers and even from some students about use of AI
• 12:45
  SPEAKER_01: in schools and the effects it is having and how to teach effectively and figure out
• 12:50
  SPEAKER_01: if students are actually learning and to help them to actually learn
• 12:54
  SPEAKER_01: and not just parrot back something that sounds plausible.
• 12:58
  SPEAKER_01: It's a real challenge for teachers.
• 13:00
  SPEAKER_00: It is a real challenge.
• 13:02
  SPEAKER_00: And then if let's say most of our knowledge is acquired through Instagram or TikTok
• 13:10
  SPEAKER_00: and We all know these algorithms work through rage baits.
• 13:14
  SPEAKER_00: What does that do to our democracy?
• 13:16
  SPEAKER_00: What does it do to our democratic literacy when all kinds of right wing
• 13:21
  SPEAKER_00: or extreme parties through these altmic preferences and
• 13:26
  SPEAKER_00: through paid mechanisms target young minds and not just young minds.
• 13:31
  SPEAKER_00: Our age groups are just as sensitive to that,
• 13:34
  SPEAKER_00: but at least we have grown up with some kind of literacy.
• 13:38
  SPEAKER_01: Yeah, really good points.
• 13:40
  SPEAKER_01: I want to go back to as far as using these different AI tools.
• 13:44
  SPEAKER_01: If you have a specific story at a time that you used one,
• 13:47
  SPEAKER_01: maybe one time when it worked well for you and another time when
• 13:50
  SPEAKER_01: the tool did not work so well for you.
• 13:53
  SPEAKER_00: In the early days,
• 13:54
  SPEAKER_00: I was using ChatGPT and I did like it as a sparing partner
• 14:01
  SPEAKER_00: because sometimes it gave me language for certain concepts.
• 14:04
  SPEAKER_00: It helped me also deepen my thinking.
• 14:08
  SPEAKER_00: Another time I was researching for a paper and gave me all these wonderful quotes.
• 14:13
  SPEAKER_00: They were all wrong.
• 14:15
  SPEAKER_00: Luckily,
• 14:16
  SPEAKER_00: I noticed because I fact checked those papers,
• 14:18
  SPEAKER_00: but I closely know people that fell prey to that in the early days and had
• 14:28
  SPEAKER_00: a completely wrong list of citations.
• 14:31
  SPEAKER_01: I know you don't use the tools now as much as you used to for various reasons.
• 14:35
  SPEAKER_01: I think you said that.
• 14:37
  SPEAKER_01: But obviously learning to fact check,
• 14:39
  SPEAKER_01: I hope your friends are doing that now and they've learned that they know
• 14:43
  SPEAKER_01: the tools are better and they may hallucinate less that they still need
• 14:46
  SPEAKER_01: to always verify.
• 14:48
  SPEAKER_00: Yes, absolutely.
• 14:49
  SPEAKER_00: Absolutely.
• 14:50
  SPEAKER_00: One time a student of mine,
• 14:53
  SPEAKER_00: two years ago,
• 14:55
  SPEAKER_00: he handed in a paper which was terribly prompted and he sounded like an old lady
• 15:02
  SPEAKER_00: who is a yoga teacher.
• 15:04
  SPEAKER_00: And I had to read through 10 pages of that gibberish.
• 15:07
  SPEAKER_00: And I told him because he was bright in class.
• 15:10
  SPEAKER_00: I mean, very bright.
• 15:12
  SPEAKER_00: I told him, okay, I have received your email, your paper.
• 15:17
  SPEAKER_00: I will have to fail you unless tomorrow I see something worthwhile in my inbox.
• 15:26
  SPEAKER_00: And then this guy wrote the best paper of his whole grade.
• 15:31
  SPEAKER_00: It wasn't perfect English, but it was wonderful.
• 15:34
  SPEAKER_00: And I think he never would have found out, neither would I.
• 15:38
  SPEAKER_00: And there's pleasure in that.
• 15:39
  SPEAKER_00: So for me, that's definitely a nice memory.
• 15:44
  SPEAKER_01: Well, that sounds good.
• 15:46
  SPEAKER_01: So you've talked about caution about using AI tools.
• 15:50
  SPEAKER_01: Are there any specific things that you would say I will never use AI to do this.
• 15:57
  SPEAKER_01: And examples of when and why you would choose not to use AI for that purpose.
• 16:01
  SPEAKER_00: So people I have a personal relationship with, I would never write them with AI.
• 16:07
  SPEAKER_00: I have received personal messages.
• 16:10
  SPEAKER_00: There have been a few conflictual moments and people hid behind written messages.
• 16:16
  SPEAKER_00: And I knew from them that they were heavy users and
• 16:20
  SPEAKER_00: the messages really I'd say they had a bit of a crazy undertone because of
• 16:27
  SPEAKER_00: the confirmation bias.
• 16:29
  SPEAKER_00: And I felt slightly insulted that the person didn't use their own minds to do that.
• 16:35
  SPEAKER_00: But yeah, so I love to use AI formal emails to some kind of administrative body.
• 16:44
  SPEAKER_00: That's wonderful.
• 16:46
  SPEAKER_00: But those relational things,
• 16:49
  SPEAKER_00: I personally feel it crosses the line, but this is a very subjective opinion.
• 16:54
  SPEAKER_01: Of course, yeah, that makes sense.
• 16:56
  SPEAKER_01: Have you ever tried any of the AI tools for music or for generating images or
• 17:02
  SPEAKER_01: for videos or any of those kinds of things other than words?
• 17:06
  SPEAKER_00: A friend of mine, when I had a funny situation, generated a song for me on SOMO AI.
• 17:13
  SPEAKER_00: I think it's called that.
• 17:14
  SPEAKER_00: That was fun.
• 17:15
  SPEAKER_00: I've seen friends generate videos,
• 17:17
  SPEAKER_00: so I do see some potential in that, but I personally haven't done that yet.
• 17:21
  SPEAKER_01: Yeah.
• 17:22
  SPEAKER_01: Okay.
• 17:23
  SPEAKER_01: I was just curious as to whether you had any concerns about that, any reservations?
• 17:28
  SPEAKER_00: I do, in the sense that,
• 17:30
  SPEAKER_00: for instance,
• 17:31
  SPEAKER_00: now there's a German court case where these AIs have used original content
• 17:37
  SPEAKER_00: and reproduced it.
• 17:39
  SPEAKER_00: GEMA, that's the German ruling body for that,
• 17:42
  SPEAKER_00: filed a case and they have won, but now which AI was it has appealed.
• 17:47
  SPEAKER_00: So that's really interesting that Germany is taking a stance there.
• 17:51
  SPEAKER_00: You and me,
• 17:51
  SPEAKER_00: we've already spoken about how a lot of artists didn't have their work protected
• 17:56
  SPEAKER_00: and it's just reproduced.
• 17:58
  SPEAKER_00: In one of those German rulings,
• 18:00
  SPEAKER_00: it was interesting that it's okay to be inspired by an idea but it's different
• 18:05
  SPEAKER_00: to just reproduce the whole thing so I think AI had gotten inspired by a photo of
• 18:11
  SPEAKER_00: a photographer but there the court ruled it's not an infringement
• 18:16
  SPEAKER_00: because it's just inspired by it so there's still quite a bit
• 18:19
  SPEAKER_00: of gray area there's still a bit of room for interpretation and I'm pretty excited
• 18:27
  SPEAKER_00: about where this is going to go Yeah,
• 18:30
  SPEAKER_01: I think it's going to take quite a while before all of this works through the courts
• 18:33
  SPEAKER_01: and all the different countries and hopefully we end up somewhere near
• 18:37
  SPEAKER_01: the same places because material crosses boundaries easily.
• 18:42
  SPEAKER_00: That's one thing.
• 18:44
  SPEAKER_00: And the other thing is I feel that quite a few of these AI companies have
• 18:49
  SPEAKER_00: that premise of rather saying sorry than asking for permission.
• 18:52
  SPEAKER_00: So a lot of boundaries are crossed.
• 18:54
  SPEAKER_00: And what do we do once they are crossed?
• 18:57
  SPEAKER_00: How do we handle that?
• 18:58
  SPEAKER_00: And same with the data centers.
• 19:01
  SPEAKER_00: So what do we do when we have a water shortage in certain areas?
• 19:04
  SPEAKER_00: So there I'm quite cautious.
• 19:07
  SPEAKER_00: And I think we are going to face challenging times.
• 19:11
  SPEAKER_01: As far as the sources and whether or not there's infringement,
• 19:14
  SPEAKER_01: that's one of the big concerns where companies get the data that they use
• 19:19
  SPEAKER_01: for training these systems, whether they acquire it legally or ethically.
• 19:23
  SPEAKER_01: And if they respect the copyright,
• 19:25
  SPEAKER_01: another aspect that comes up is not just the infringement,
• 19:28
  SPEAKER_01: but whether it's unfair competition.
• 19:31
  SPEAKER_01: So we're seeing lawsuits in the US on that basis and not just on the theft of
• 19:36
  SPEAKER_01: the original materials.
• 19:37
  SPEAKER_01: I'm wondering how you feel about this concept they call the three C's,
• 19:41
  SPEAKER_01: which is that creative people should be entitled to consent,
• 19:45
  SPEAKER_01: credit, and compensation for the use of their materials for AI.
• 19:50
  SPEAKER_01: Some people feel that as long as it's for the greater good,
• 19:53
  SPEAKER_01: and I'll put that in quotes, that it's okay for their material to be used.
• 19:58
  SPEAKER_01: But other people say, no, that's my life's work and I need to make a living from it.
• 20:04
  SPEAKER_01: I deserve to have the choice and I deserve to be credited and compensated if my work is used.
• 20:09
  SPEAKER_01: Where do you come down on that thought?
• 20:12
  SPEAKER_00: I absolutely believe in the three C's.
• 20:14
  SPEAKER_00: I didn't know they were called that way, but it's the bare minimum to ask for consent.
• 20:20
  SPEAKER_00: It's your intellectual property and there are rules and regulations around that.
• 20:24
  SPEAKER_00: Just because AI has kind of swamped the market and just did things,
• 20:28
  SPEAKER_00: it's not AI that did those things,
• 20:30
  SPEAKER_00: but the companies behind that, people need to be protected.
• 20:33
  SPEAKER_00: Again, in this David and Goliath kind of situation,
• 20:36
  SPEAKER_00: a small Etsy creator doesn't have the same power as OpenAI, for instance.
• 20:44
  SPEAKER_00: The other question is, how can we even find out?
• 20:46
  SPEAKER_00: I mean, some people randomly find their work reproduced,
• 20:52
  SPEAKER_00: but I think there are numerous people who haven't found their work being reproduced
• 20:57
  SPEAKER_00: yet.
• 20:57
  SPEAKER_01: Do you know of any cases where your work,
• 21:00
  SPEAKER_01: either personally,
• 21:02
  SPEAKER_01: your writing or other materials or anything from Finding Sustainia has been used
• 21:07
  SPEAKER_01: without your consent?
• 21:09
  SPEAKER_00: It has, but I don't think it was AI.
• 21:13
  SPEAKER_00: But yet that happens in the sustainability world as well.
• 21:18
  SPEAKER_00: And I'm each time shocked.
• 21:20
  SPEAKER_01: Yeah, plagiarism, people copying and stealing each other's work.
• 21:24
  SPEAKER_01: We've been seeing that on LinkedIn and other places for years,
• 21:27
  SPEAKER_01: long before generative AI really took over.
• 21:29
  SPEAKER_01: Maybe makes it easier in more and more common but it didn't really start with that
• 21:34
  SPEAKER_01: and it's the same thing with theft of our personal data and having it sold
• 21:38
  SPEAKER_01: to data brokers that's been going on for decades so that's not really new it's just
• 21:42
  SPEAKER_01: that AI amplifies it and makes it perhaps more serious because once our data goes
• 21:47
  SPEAKER_01: into an AI system for training it really isn't ever going to come back out
• 21:52
  SPEAKER_01: and that's an issue for instance some of the principles in Germany
• 21:56
  SPEAKER_01: under GDPR people have the right
• 21:57
  SPEAKER_01: to be forgotten that's pretty hard actually once AI gets in
• 22:01
  SPEAKER_01: the loop
• 22:03
  SPEAKER_00: It is very hard.
• 22:04
  SPEAKER_00: And then there are these cases where a woman didn't consent to her videos
• 22:10
  SPEAKER_00: or her face being suddenly used for deepfake porn, for instance.
• 22:14
  SPEAKER_00: I mean, just imagine how violated we would feel if that happened to us.
• 22:20
  SPEAKER_00: And these things need to be heavily regulated.
• 22:23
  SPEAKER_00: It just shouldn't be possible that AI can play around with your person without your consent.
• 22:32
  SPEAKER_00: So there's a lot of gray zones there that in my opinion shouldn't be gray zones.
• 22:36
  SPEAKER_00: Yeah.
• 22:36
  SPEAKER_01: Do you know any company that you gave your data to that told you upfront
• 22:40
  SPEAKER_01: that they might use it for AI or machine learning?
• 22:44
  SPEAKER_00: No, not that I can think of.
• 22:46
  SPEAKER_00: Do you know any company?
• 22:49
  SPEAKER_01: Well, I tended to change the terms and conditions after we signed up.
• 22:52
  SPEAKER_01: I signed up for a Facebook account many years ago and back then it was not
• 22:56
  SPEAKER_01: a question, but then they surreptitiously changed their terms and conditions.
• 23:00
  SPEAKER_01: They were hiding it.
• 23:02
  SPEAKER_01: And then the news started to come out.
• 23:04
  SPEAKER_01: We find that people are often becoming distrustful of these companies and
• 23:09
  SPEAKER_01: with good reason because they are doing things that we didn't agree to.
• 23:13
  SPEAKER_01: People do feel violated and it's not being treated with respect by these companies.
• 23:20
  SPEAKER_01: Like you said, they feel like they can get away with it.
• 23:23
  SPEAKER_01: And so they try to, and that's just not ethical behavior.
• 23:27
  SPEAKER_00: Absolutely.
• 23:27
  SPEAKER_00: I highly recommend two Netflix documentaries,
• 23:30
  SPEAKER_00: and that was before AI even rolled out at scale,
• 23:34
  SPEAKER_00: and that is The Big Hack and Social Dilemma.
• 23:38
  SPEAKER_00: And yeah, it's hard to even assume that meta has any good intentions.
• 23:44
  SPEAKER_00: Not many boundaries there.
• 23:47
  SPEAKER_00: Yeah.
• 23:47
  SPEAKER_01: The Social Dilemma I've heard of, but not The Big Hack.
• 23:49
  SPEAKER_01: I'm going to have to go look that one up.
• 23:51
  SPEAKER_00: I think you'll enjoy it.
• 23:53
  SPEAKER_01: So when you think about AI adoption,
• 23:57
  SPEAKER_01: either at the individual personal level or at the level of a company that is looking
• 24:01
  SPEAKER_01: to improve their business with AI.
• 24:03
  SPEAKER_01: What role do you see for governance in those adoptions?
• 24:08
  SPEAKER_00: For me,
• 24:09
  SPEAKER_00: governance is much bigger than compliance and I feel AI is often seen through
• 24:15
  SPEAKER_00: a compliance lens and it's really about designing organizations
• 24:20
  SPEAKER_00: that create good incentives, clear accountability and responsible decision making.
• 24:26
  SPEAKER_00: And for me,
• 24:27
  SPEAKER_00: good Governance means anticipating these second and third order effects instead
• 24:32
  SPEAKER_00: of just reacting once problems appear.
• 24:35
  SPEAKER_00: And I feel we are still very reactive, even though quite a few things are foreseeable.
• 24:41
  SPEAKER_00: There are a lot of unknown unknowns, but quite a
• 24:44
  SPEAKER_01: few things can be extrapolated.
• 24:48
  SPEAKER_01: How do you see the issues within governments being responsible,
• 24:51
  SPEAKER_01: but also how does it affect leadership
• 24:54
  SPEAKER_01: and also you had mentioned organizational design is a challenge and I would like
• 24:58
  SPEAKER_01: to hear more of your thoughts about that.
• 25:00
  SPEAKER_00: So in my opinion organizations still lack a lot of literacy.
• 25:06
  SPEAKER_00: So I saw one study where an organization invested so much money and
• 25:10
  SPEAKER_00: then their staff isn't really adapting it or is adapting it wrongly
• 25:15
  SPEAKER_00: and AI is often treated as an IT project.
• 25:18
  SPEAKER_00: I don't know what your experience is,
• 25:19
  SPEAKER_00: but they quite naively believe that AI is just going to solve
• 25:24
  SPEAKER_00: so many different problems.
• 25:25
  SPEAKER_00: Whereas an organization, it's a lot about coherence.
• 25:29
  SPEAKER_00: It's a lot about creating the kinds of structures that you're employees will thrive
• 25:34
  SPEAKER_00: and can do their work well.
• 25:36
  SPEAKER_00: So AI would ideally replace repetitive tasks,
• 25:42
  SPEAKER_00: but often now it's coming on top of a lot of different tasks
• 25:46
  SPEAKER_00: and people are quite frazzled.
• 25:48
  SPEAKER_00: Again, I don't know what your impression is, but it's a lot about leadership, trust.
• 25:52
  SPEAKER_00: So all those things still matter very much when we think of AI.
• 25:56
  SPEAKER_01: Yeah, there have been some really interesting recent studies about the impact
• 26:00
  SPEAKER_01: that AI is having on people.
• 26:02
  SPEAKER_01: The claims are always that it's meant to enhance productivity and it takes away some
• 26:07
  SPEAKER_01: of the boring activity.
• 26:08
  SPEAKER_01: But what they're finding is that people's brains are actually more stressed
• 26:12
  SPEAKER_01: because doing some of the boring stuff used to give their minds a break.
• 26:17
  SPEAKER_01: And now they're not getting those breaks because they're just jumping
• 26:20
  SPEAKER_01: from one heavy thinking activity into another.
• 26:23
  SPEAKER_01: And people were feeling more burned out from that.
• 26:27
  SPEAKER_01: They're being more productive, but they're also mentally exhausted at the end of the day.
• 26:31
  SPEAKER_01: Whereas they used to have these little mental breaks from doing some of
• 26:34
  SPEAKER_01: the other tasks that now AI is handling.
• 26:38
  SPEAKER_00: And some leaders actually use it like a little child in the sense that they get
• 26:43
  SPEAKER_00: so enthusiastic and then they swamp their teams with hundreds of pages
• 26:49
  SPEAKER_00: that is AI generation.
• 26:50
  SPEAKER_00: They don't even have a distance to that work anymore.
• 26:53
  SPEAKER_00: I know of a case.
• 26:55
  SPEAKER_00: Reporting consultancy and one of the managers,
• 27:00
  SPEAKER_00: high ranking,
• 27:01
  SPEAKER_00: was just swamping their team and the team was feeling burnt out
• 27:03
  SPEAKER_00: because these reports were gibberish.
• 27:05
  SPEAKER_00: They sounded coherent, but then they had to wade through it.
• 27:10
  SPEAKER_00: Just the way that you could do that erodes trust when you act like this.
• 27:15
  SPEAKER_00: So there needs to be a mindful way.
• 27:18
  SPEAKER_00: And I don't know whether you've noticed,
• 27:19
  SPEAKER_00: but when you read AI,
• 27:21
  SPEAKER_00: sometimes your brain drifts off because it sounds so coherent
• 27:26
  SPEAKER_00: that sometimes our critical skills actually decrease as well.
• 27:32
  SPEAKER_01: Yeah.
• 27:32
  SPEAKER_01: What I hear from a lot of people is that they read it and it sounds polished,
• 27:36
  SPEAKER_01: but it's all veneer.
• 27:38
  SPEAKER_01: There's nothing solid underneath it.
• 27:40
  SPEAKER_01: There are cartoon jokes.
• 27:41
  SPEAKER_01: Someone is saying, please expand these three bullets into a full report.
• 27:45
  SPEAKER_01: And then they send off the report and then someone else says,
• 27:47
  SPEAKER_01: please take this report and distill it to bullets for me.
• 27:51
  SPEAKER_01: And then it's like playing telephone.
• 27:52
  SPEAKER_01: Do you get back the same three bullets that the person started with or not?
• 27:56
  SPEAKER_01: And using AI on both ends, it seems like it didn't really help the communications.
• 28:01
  SPEAKER_01: They would have been better to stick to the bullets in the first place and be clear
• 28:04
  SPEAKER_01: about what the intent was.
• 28:07
  SPEAKER_00: Yes.
• 28:07
  SPEAKER_00: So many organizations have an intent.
• 28:09
  SPEAKER_00: The first order is speed and cost,
• 28:13
  SPEAKER_00: but then they ignore that trust and adoption can become an obstacle.
• 28:17
  SPEAKER_00: And then the third order effect comes that it erodes resilience and company culture.
• 28:24
  SPEAKER_01: You mentioned trust.
• 28:25
  SPEAKER_01: We do see that public distrust of these AI and tech companies has been growing.
• 28:30
  SPEAKER_01: So what would you recommend as the most important thing that companies could do
• 28:35
  SPEAKER_01: to earn and keep?
• 28:37
  SPEAKER_01: Your trust as a consumer, as an employee perhaps, or as someone who's working with a company.
• 28:43
  SPEAKER_01: What would a company need to do for you to feel like you could trust them,
• 28:48
  SPEAKER_01: if that's even possible?
• 28:50
  SPEAKER_00: It is possible.
• 28:51
  SPEAKER_00: We're talking about an AI company, right?
• 28:54
  SPEAKER_00: That would go through transparency, accountability and limitations.
• 28:59
  SPEAKER_00: What you were talking about with the three C's that creators are asked that
• 29:04
  SPEAKER_00: if they make a glitch,
• 29:05
  SPEAKER_00: they communicate about it because we're talking about artificial intelligence.
• 29:09
  SPEAKER_00: Intelligence so things don't just always go as we plan it.
• 29:12
  SPEAKER_00: So there are those unknown unknowns,
• 29:14
  SPEAKER_00: but then there are also those infringements that are clearly company policy.
• 29:19
  SPEAKER_00: You know of the Reuters study that showed that
• 29:22
  SPEAKER_00: the meta AI bot had been actually trained to sexually elude and flirt with children.
• 29:31
  SPEAKER_00: And how do you own that?
• 29:32
  SPEAKER_00: Of course,
• 29:33
  SPEAKER_00: trust is going to be difficult because it was intended to do so,
• 29:36
  SPEAKER_00: but If they are not accountable by themselves, they need to be held accountable.
• 29:42
  SPEAKER_01: Absolutely.
• 29:42
  SPEAKER_01: So if you're thinking about a company that wants to embrace AI,
• 29:47
  SPEAKER_01: but actually cares about not losing their critical thinking,
• 29:50
  SPEAKER_01: not losing their human judgment, what would you recommend that they do?
• 29:57
  SPEAKER_00: It's about how do we define the intention of the tools?
• 30:01
  SPEAKER_00: So for me,
• 30:02
  SPEAKER_00: AI can generate options,
• 30:05
  SPEAKER_00: analyze information, and also challenge assumptions, which is very valuable.
• 30:09
  SPEAKER_00: But as humans,
• 30:10
  SPEAKER_00: we still need to define the values,
• 30:12
  SPEAKER_00: balance the trade-offs, and yeah, remain accountable for decisions.
• 30:18
  SPEAKER_01: So using our wetware, using our judgment.
• 30:21
  SPEAKER_00: Yes, absolutely.
• 30:23
  SPEAKER_00: And it's important that AI doesn't kind of take off in a way
• 30:29
  SPEAKER_00: that we can't follow it anymore mentally.
• 30:32
  SPEAKER_00: It's one thing that it's productive, but our brains can't handle this kind of speed.
• 30:39
  SPEAKER_00: You remember maybe 10 years ago how we were talking about how good it is for us
• 30:44
  SPEAKER_00: to chop veggies and it's not always good to, I don't know, have ready-made meals.
• 30:49
  SPEAKER_00: I think that's the same.
• 30:50
  SPEAKER_00: It's good to write on pen and paper or by yourself.
• 30:54
  SPEAKER_00: I have a horrible handwriting, but to have that human pace.
• 30:58
  SPEAKER_00: And I think as an antidote to that speed,
• 31:02
  SPEAKER_00: we need to have these conversations like we are right now doing.
• 31:06
  SPEAKER_00: We need to sit down with a book.
• 31:07
  SPEAKER_00: So that's actually also a way for me to actively protect my wet where I put my phone away.
• 31:14
  SPEAKER_00: I sit down and read and also make it a point to talk to people because these are
• 31:18
  SPEAKER_00: the things that AI can't and shouldn't replace.
• 31:21
  SPEAKER_00: But unfortunately it's not our normal anymore and we need to be very proactive
• 31:26
  SPEAKER_00: about it and protective.
• 31:29
  SPEAKER_01: Yeah, absolutely.
• 31:29
  SPEAKER_01: We need to keep building those human skills.
• 31:31
  SPEAKER_01: One of the people that I'm working with now on a collaboration for our second set
• 31:36
  SPEAKER_01: of She Writes AI books has talked
• 31:38
  SPEAKER_01: about something she has called somographic learning where Basically,
• 31:42
  SPEAKER_01: the more the body is engaged in something, the better you learn it.
• 31:46
  SPEAKER_01: And people have seen this for years,
• 31:48
  SPEAKER_01: that if you take handwritten notes in a class,
• 31:51
  SPEAKER_01: you somehow retain them better than if you type your notes.
• 31:53
  SPEAKER_01: And I found that to be true for myself as well.
• 31:55
  SPEAKER_01: Even in meetings, I would often handwrite notes to help me remember them.
• 32:00
  SPEAKER_01: And so when we get to note takers doing even that part of it for us,
• 32:04
  SPEAKER_01: instead of us typing them in, we have a note taker that's capturing them.
• 32:08
  SPEAKER_01: It's a little more distant and people don't retain it as well.
• 32:12
  SPEAKER_01: I think those dynamics are quite interesting.
• 32:15
  SPEAKER_01: It's like you said,
• 32:15
  SPEAKER_01: some people find that reading a physical book they retain better than if they read
• 32:20
  SPEAKER_01: on a screen.
• 32:20
  SPEAKER_01: So I think there's a lot of...
• 32:22
  SPEAKER_00: Yes.
• 32:23
  SPEAKER_00: And that's why we do need ambition, but we also need humility.
• 32:26
  SPEAKER_00: And also we need to slow down at times and reflect and try
• 32:30
  SPEAKER_00: to integrate those reflections into the technological process
• 32:35
  SPEAKER_00: because technology is evolving incredibly quickly.
• 32:40
  SPEAKER_00: And we as a society, as organizations, we just take longer to adapt.
• 32:45
  SPEAKER_00: And that's also why governance matters so much.
• 32:49
  SPEAKER_01: So as we think about living in this age of AI,
• 32:52
  SPEAKER_01: what skills do you think become the most valuable?
• 32:57
  SPEAKER_00: That's such a beautiful question.
• 33:00
  SPEAKER_00: So definitely curiosity and sometimes just holding uncertainty,
• 33:07
  SPEAKER_00: giving ourselves space to ponder about something instead of looking
• 33:10
  SPEAKER_00: for quick answers.
• 33:12
  SPEAKER_00: We need to also protect our critical judgment and our democratic literacy, fact check.
• 33:20
  SPEAKER_00: Systems thinking is very important to just understand that your individual behaviors
• 33:26
  SPEAKER_00: and correlation with a bigger picture and who you want to be in that bigger picture,
• 33:31
  SPEAKER_00: even though you're also small, but still it matters.
• 33:35
  SPEAKER_00: Then we also need to learn about ethical reasoning, communication.
• 33:40
  SPEAKER_00: I think it's also very underrated, also often outsourced to AI.
• 33:45
  SPEAKER_00: Emotional intelligence,
• 33:47
  SPEAKER_00: which also means we need to have tools,
• 33:49
  SPEAKER_00: how to have a constructive, positive conflict.
• 33:53
  SPEAKER_00: And then again, asking better questions.
• 33:55
  SPEAKER_00: You're very good at good questions.
• 33:57
  SPEAKER_01: Thank you.
• 33:58
  SPEAKER_01: I think we share that.
• 34:00
  SPEAKER_00: Yes.
• 34:00
  SPEAKER_00: And information is so abundant right now.
• 34:03
  SPEAKER_00: We are swamped on all kinds of levels and hence it's important
• 34:08
  SPEAKER_00: to protect our judgment and our discernment.
• 34:12
  SPEAKER_00: And all those attributes actually sound quite boring.
• 34:14
  SPEAKER_00: I remember when I was a child and they came with stuff like courage
• 34:18
  SPEAKER_01: and honor and
• 34:19
  SPEAKER_01: the different virtues,
• 34:21
  SPEAKER_01: but looking
• 34:24
  SPEAKER_00: at our time right now, I think they are more relevant than ever.
• 34:28
  SPEAKER_01: I appreciate you making time to come on this interview with me, Shanta.
• 34:33
  SPEAKER_01: Is there anything else you'd like to share with your audience?
• 34:35
  SPEAKER_01: If you could leave leaders with one question about AI,
• 34:39
  SPEAKER_01: what would it be or what would you want them to think about?
• 34:43
  SPEAKER_01: Let me give it a think.
• 34:44
  SPEAKER_00: The question I would like the audience to leave with is instead
• 34:48
  SPEAKER_00: of asking what AI can do,
• 34:51
  SPEAKER_00: rather ask what kind of organization and ultimately what kind
• 34:55
  SPEAKER_00: of society do we want AI to help us build?
• 35:00
  SPEAKER_01: Well, now there's a great question.
• 35:02
  SPEAKER_00: Thank you.
• 35:03
  SPEAKER_00: And it's even late in Germany.
• 35:04
  SPEAKER_01: I appreciate you staying up late for this call.
• 35:07
  SPEAKER_01: Anything you want to tell our audience about where to find you, to find Sustainia?
• 35:13
  SPEAKER_00: Yes, so you can find us under sustenia.com and on socials also under FindingSustenia.
• 35:20
  SPEAKER_00: We also have a substack.
• 35:22
  SPEAKER_00: Maybe you could link it under the podcast.
• 35:25
  SPEAKER_00: And yes,
• 35:27
  SPEAKER_00: please stay in touch with each other,
• 35:29
  SPEAKER_00: speak about things, speak about your frictions, about your fascination with AI.
• 35:34
  SPEAKER_00: I think staying in touch with humans and protecting our wetware are very delicious
• 35:41
  SPEAKER_00: and important tasks.
• 35:43
  SPEAKER_00: Sounds great.
• 35:44
  SPEAKER_00: Thank you so much, Shonda.
• 35:45
  SPEAKER_00: Take care.
• 35:46
  SPEAKER_00: Thank you, Karen.
• 35:49
  SPEAKER_01: About this interview series and newsletter.
• 35:52
  SPEAKER_01: This post is part of our AI6P interview series on AI software and wetware.
• 35:59
  SPEAKER_01: It showcases how real people around the world are using their wetware, brains,
• 36:03
  SPEAKER_01: and human intelligence with AI-based software tools who are being affected by AI.
• 36:11
  SPEAKER_01: And we're all being affected by AI nowadays in our daily lives, perhaps more than we realize.
• 36:17
  SPEAKER_01: For some examples, see post, but I don't use AI.
• 36:21
  SPEAKER_01: We want to hear from a diverse pool of people worldwide in a variety of roles.
• 36:26
  SPEAKER_01: If you're interested in being a featured interview guest,
• 36:29
  SPEAKER_01: anonymous or with credit, please get in touch.
• 36:36
  SPEAKER_01: All new posts are free to read and listen to.
• 36:46
  SPEAKER_01: To automatically receive new AI6P posts and support our work, consider becoming a subscriber.
• 36:52
  SPEAKER_01: It's free.