Latest / kill switch / workslop: how AI is making you hate your coworkers
Transcript
- 0:09Speaker 1: One of the first examples that I saw was a
- 0:13Speaker 1: really long memo with this style that we talk about
- 0:17Speaker 1: as purple prose.
- 0:18Speaker 2: It's like when people.
- 0:19Speaker 1: Just use really flowery language that's elaborate and long.
- 0:25Speaker 3: Kate Niedhoffer is a social psychologist and the vice president
- 0:28Speaker 3: of Betterop Labs, which recently conducted a study about a
- 0:31Speaker 3: new phenomenon in the workplace AI WORKSLOP.
- 0:34Speaker 1: We ask people to share examples of workslops that they've received.
- 0:38Speaker 1: They'll tell us like, there's this weirdness, like there's something
- 0:42Speaker 1: off about this, and feeling so confused and unsure what
- 0:45Speaker 1: the content is really about. But then it's like I
- 0:49Speaker 1: don't know what to do about this. I don't know
- 0:51Speaker 1: if I should just redo it myself, start from scratch,
- 0:55Speaker 1: tell the person, ask the person. Sometimes there's a power
- 0:59Speaker 1: dynamic that makes it really complicated to even engage, and
- 1:03Speaker 1: so it leaves people sort of paralyzed. It's like I
- 1:06Speaker 1: don't know what this is, what it means, or what
- 1:08Speaker 1: to do about it, and I still have to do
- 1:10Speaker 1: the work.
- 1:11Speaker 3: What Kate's study found is that contrary to what AI
- 1:14Speaker 3: companies have been promising, AI is creating more work for
- 1:17Speaker 3: people we're having to spend time reading emails written by
- 1:19Speaker 3: chat GPT, watching automatically generated PowerPoint presentations, and dealing with
- 1:24Speaker 3: computer code that looks okay but doesn't actually run properly. Meanwhile,
- 1:29Speaker 3: Mark Zuckerberg is telling his workers to use AI to
- 1:32Speaker 3: work five times faster, and a bunch of companies are
- 1:35Speaker 3: still laying off employees under the assumption that they can
- 1:38Speaker 3: save money by just using AI to do that work.
- 1:41Speaker 1: What we're doing is slashing the human and eliminating that
- 1:46Speaker 1: whole productivity potential. I think it's backfiring because you're having
- 1:51Speaker 1: a well being tax on the people who stay, who
- 1:53Speaker 1: have lost their peers, who are taking on all of
- 1:56Speaker 1: this work. It's just a compounding negative effect.
- 2:02Speaker 3: This study went kind of viral and made some headlines,
- 2:04Speaker 3: and if you just read the headlines, it can seem
- 2:07Speaker 3: like the results prove that AI shouldn't be used in
- 2:09Speaker 3: the workplace at all, But that wasn't really the point
- 2:12Speaker 3: of the story. Betterup's main thing is that they give
- 2:15Speaker 3: advice to companies. Usually that advice isn't just given out
- 2:18Speaker 3: for free in public. Companies pay money for this insight.
- 2:22Speaker 3: So my interview with Kate was an opportunity to see
- 2:25Speaker 3: what kind of conversations are actually happening behind the scenes
- 2:28Speaker 3: at workplaces that are experimenting with AI, the kind of
- 2:31Speaker 3: conversations that might also be happening where you work. Kaleidoscope
- 2:44Speaker 3: and iHeart podcasts.
- 2:46Speaker 4: This is kill switch. I'm dexter Thomas, I'm sorry, goodbye.
- 3:29Speaker 5: How would you define workslap workslop is AI generated text
- 3:35Speaker 5: that looks like it completes a task, but.
- 3:40Speaker 1: Upon further inspection reveals that it lacks context and all
- 3:45Speaker 1: the cognitive effort that is really necessary to complete a task.
- 3:50Speaker 3: So, in other words, it's when you get sent an
- 3:52Speaker 3: email or a presentation that's very clearly written by chat GBT,
- 3:56Speaker 3: when there's a bunch of M dashes and bullet points,
- 3:58Speaker 3: flowery language phrases like delve into the stuff that at
- 4:02Speaker 3: first glanced kind of looks like a well thought out
- 4:05Speaker 3: piece of work, but if you keep reading, it's obviously robotic.
- 4:09Speaker 1: We see a lot of examples of emails, reports, code,
- 4:14Speaker 1: a lot of like vibe coded code, even apps. So
- 4:18Speaker 1: people talked about the creation of prototypes, for example, that
- 4:22Speaker 1: are based on prompts, and so they're sort of brittle
- 4:24Speaker 1: experiences that don't really have substance underneath them.
- 4:28Speaker 3: Where did this phrase work slob come.
- 4:30Speaker 1: From My colleague Jeff Hancock at the Stanford Social Media Lab.
- 4:34Speaker 1: And we had been studying the ways that managers make
- 4:39Speaker 1: decisions about delegating tasks to humans versus AI, and we
- 4:44Speaker 1: knew about AI slop, you know, that already existed, and
- 4:47Speaker 1: so we were thinking about, like, but what is it
- 4:49Speaker 1: in the workplace and why is it so different? So
- 4:52Speaker 1: Jeff came up with that first definition of AI generated
- 4:55Speaker 1: content that fulfills the appearance of completing the task, but
- 4:59Speaker 1: it doesn't actually have this like substantive cognitive work or
- 5:02Speaker 1: the decision making that a task really requires. And so
- 5:06Speaker 1: we went out with this definition we started to study it.
- 5:10Speaker 3: Kate and her team worked with Stanford Social Media Lab
- 5:12Speaker 3: to study this. They surveyed over one thousand workers across
- 5:15Speaker 3: fields from finance to medicine to government.
- 5:18Speaker 1: We start by kind of like funneling them into it
- 5:21Speaker 1: by saying, like, we have some questions about your experience
- 5:24Speaker 1: with AI at work. We explain what AI is so
- 5:27Speaker 1: we're all on the same page, and then we ask
- 5:29Speaker 1: them how often do you use AI at work? And
- 5:31Speaker 1: we ask if people in their company have required them
- 5:35Speaker 1: to use AI. As you know, many people are mandated
- 5:38Speaker 1: to use AI, so we want to get a sense
- 5:40Speaker 1: of what their policy is. And then we start asking
- 5:43Speaker 1: questions about how often they send or share work with
- 5:46Speaker 1: colleagues that uses AI to help produce it. And then
- 5:50Speaker 1: we say, how much of the AI generated work that
- 5:53Speaker 1: you send to colleagues do you think is actually unhelpful,
- 5:55Speaker 1: low effort, or low quality?
- 5:57Speaker 2: And people readily admit to that.
- 6:01Speaker 1: And then we say, well, how much of the AI
- 6:04Speaker 1: generated work that you send a colleagues do you think
- 6:06Speaker 1: is helpful and high quality? So you know, we have
- 6:09Speaker 1: to make sure that we're not being biased. And then
- 6:11Speaker 1: we say our most important question to give us the
- 6:14Speaker 1: prevalence is in your job, have you received work content
- 6:17Speaker 1: in the last month that you believe is AI generated
- 6:20Speaker 1: that looks like it completes a task at work, but
- 6:23Speaker 1: is actually unhelpful, low quality, or seems like the sender
- 6:28Speaker 1: didn't put in enough effort.
- 6:29Speaker 3: So you asked that question, which is essentially your definition
- 6:33Speaker 3: of WORKSLOP without actually using it for them so as
- 6:35Speaker 3: not to bias them. How many people answer yes.
- 6:38Speaker 2: To that question forty percent?
- 6:41Speaker 3: Forty percent in the last month. That's a really high number.
- 6:45Speaker 3: I can only imagine that's going.
- 6:46Speaker 1: To go up.
- 6:47Speaker 2: Yeah, I agree.
- 6:47Speaker 1: It's something that we're really interested in tracking. We also
- 6:52Speaker 1: wanted to know how frequently people experience it. So if
- 6:56Speaker 1: you think that that's the prevalence that forty percent of
- 6:58Speaker 1: people have experienced it, then how much of the work
- 7:01Speaker 1: that they receive do they think fits this description? And
- 7:05Speaker 1: it's about fifteen percent. So fifteen percent of the work
- 7:09Speaker 1: that you receive is AI generated work that looks like
- 7:14Speaker 1: it completes the task but really doesn't.
- 7:19Speaker 3: Okay, so this sounds a little annoying, but aside from
- 7:26Speaker 3: the mild annoying, so the deep annoying somebody might feel
- 7:28Speaker 3: from getting an email that pretty clearly is written by Chadgibt,
- 7:32Speaker 3: we're seeing a presentation that was written by chadgpt. What
- 7:35Speaker 3: else are the effects we should.
- 7:36Speaker 4: Think about here?
- 7:37Speaker 1: So it's a lot of time wasted, that's the first thing.
- 7:42Speaker 1: It has an emotional impact that's pretty strongly negative emotions
- 7:46Speaker 1: like being annoyed, frustrated.
- 7:49Speaker 2: You've been confused.
- 7:51Speaker 1: And then I think the most insidious impact of all
- 7:55Speaker 1: is this interpersonal judgment and evaluation that you think that
- 8:00Speaker 1: the person who sent you the workslop is less capable,
- 8:06Speaker 1: less creative, less trustworthy. It's like you're immediately casting judgment
- 8:12Speaker 1: on your ability to work with this person over time
- 8:15Speaker 1: and to collaborate effectively. And that's so important. That's the
- 8:19Speaker 1: most important thing in the workplace is that we believe
- 8:22Speaker 1: that our colleagues are competent, capable people with shared goals
- 8:25Speaker 1: that we can align with and do our best work.
- 8:28Speaker 3: So the respondence to your study, you're telling you this
- 8:30Speaker 3: that when they get workslop from somebody, they don't trust
- 8:33Speaker 3: that person anymore.
- 8:34Speaker 1: They trust them less.
- 8:36Speaker 3: Yeah, you know, so I've gotten pitches asking me to
- 8:41Speaker 3: cover things. So it's hey, dexter, I've seen that you
- 8:45Speaker 3: covered X. Here is this thing that I do, or
- 8:48Speaker 3: here's this thing that my client does. And it was
- 8:50Speaker 3: always in the same format. It's like a three paragraph format,
- 8:54Speaker 3: and it would be a brief sentence telling me how
- 8:57Speaker 3: cool I am, and then a couple of paragraphs telling
- 9:02Speaker 3: me what their client's product or project was, and the
- 9:06Speaker 3: third paragraph telling me why it's so interesting in giving
- 9:10Speaker 3: me suggestions for how I could cover it. And I
- 9:13Speaker 3: was seeing the same format over and over and over again.
- 9:16Speaker 3: And that made me not only dislike the PR person,
- 9:19Speaker 3: but dislike the person who I'm supposed to be covering it.
- 9:22Speaker 3: So I start to think, whoever this person is, I
- 9:26Speaker 3: don't want to have anything to do with them because
- 9:28Speaker 3: they're having some PR person who's just running their stuff
- 9:30Speaker 3: through perhaps chat Gibt, perhaps Claude, whatever, and so I
- 9:34Speaker 3: start to Yeah, I start to have a pretty negative
- 9:37Speaker 3: opinion totally.
- 9:38Speaker 1: I think that's the first phase of it. It's like
- 9:40Speaker 1: you recognize some linguistic cues that seem repetitive or template,
- 9:46Speaker 1: and like, to me, I experience it as so deceptive.
- 9:49Speaker 2: There's something first.
- 9:51Speaker 1: That's empty about the text and hard to read. But
- 9:54Speaker 1: then that second experience, which is so emotional. It's like
- 9:57Speaker 1: I'm confused, I'm frustrated, I'm annoying, angry, I feel deceived.
- 10:02Speaker 1: And then there's like that third process, that interpersonal process
- 10:05Speaker 1: that kicks in and you're like, I don't want to
- 10:07Speaker 1: work with this person, I don't trust this person, I
- 10:10Speaker 1: don't like this person in your case, and that's when
- 10:13Speaker 1: we saw those results in our study. That's when I
- 10:16Speaker 1: was like, oh God, we're in for something bad here.
- 10:20Speaker 3: The feeling of anger or frustration that you get when
- 10:22Speaker 3: someone dumps a bunch of AI generated slop on your desk.
- 10:26Speaker 3: This is what Kate would call the emotional tax of
- 10:28Speaker 3: work slop. About half the people surveyed said that when
- 10:31Speaker 3: they receive workslop from a colleague, their opinion of that
- 10:34Speaker 3: colleague changed, and at interpersonal level, this is obviously a
- 10:38Speaker 3: bad thing, but this is also the sort of thing
- 10:40Speaker 3: that a company would want to avoid because it translates
- 10:44Speaker 3: into less money for the company.
- 10:46Speaker 1: People are spending just under two hours on each incent
- 10:51Speaker 1: of workslop, and so if you think about you know,
- 10:55Speaker 1: the forty percent of people who have experienced it, the
- 10:58Speaker 1: rate with which they experience and set it can cost
- 11:01Speaker 1: up to nine million dollars for an organization. So it's
- 11:06Speaker 1: a very very costly phenomenon right now. In addition to
- 11:10Speaker 1: the cost of using these tools.
- 11:14Speaker 3: The MIT Media Lab recently reported that despite thirty to
- 11:18Speaker 3: forty billion dollars being invested in the generative AI, ninety
- 11:22Speaker 3: five percent of organizations are getting zero return on that investment.
- 11:26Speaker 3: And on top of that, Kate's study suggests that it's
- 11:29Speaker 3: actually even worse. A lot of businesses are losing money
- 11:33Speaker 3: to work slab so two questions, why is work slab
- 11:37Speaker 3: happening and whose fault is it? The second one you
- 11:40Speaker 3: might think, you know, the first one that answers a
- 11:43Speaker 3: little deeper, that's after the break So is there an
- 11:56Speaker 3: age demographic that's more likely to be getting or receiving
- 12:01Speaker 3: WORKSLOP in your study?
- 12:03Speaker 1: I don't have a demographic breakdown right now. It's something
- 12:06Speaker 1: that we're working on. We know from our previous research
- 12:10Speaker 1: that older people trust AI more, and we know that
- 12:14Speaker 1: trust in AI, especially when a sort of blind trust,
- 12:17Speaker 1: can lead to this over reliance and can produce more WORKSLOP.
- 12:22Speaker 3: So that right there is really interesting because the tendency
- 12:25Speaker 3: with any kind of new technology, with almost any new
- 12:28Speaker 3: technology in the past, we've expected for young people to
- 12:32Speaker 3: be the early adopters, and that I think societally we
- 12:34Speaker 3: start to blame young people for all of the ills
- 12:37Speaker 3: that that new technology brings. But what you're saying about
- 12:40Speaker 3: trust in AIS is really interesting.
- 12:43Speaker 1: So I think there is something really interesting happening with
- 12:46Speaker 1: age here. You know, agency understanding how to use the
- 12:50Speaker 1: tools in a discerning way, knowing when to use them,
- 12:54Speaker 1: how to use them. That's a really good positive predictor
- 12:57Speaker 1: of not producing WORKSLOP, and so it's really possible that
- 13:02Speaker 1: maybe a more tech native person would have higher agency
- 13:06Speaker 1: and be less likely to over rely on the tools.
- 13:10Speaker 1: There's this really interesting distinction and the way that people
- 13:12Speaker 1: use AI. We call it the pilot mindset. So people
- 13:15Speaker 1: who are high agency, high optimism in the way that
- 13:19Speaker 1: they use AI, and we see that pilots are not
- 13:24Speaker 1: producing as much WORKSLOP as the inverse, which is passengers.
- 13:29Speaker 1: Passengers are low agency, low optimism. They're either not using
- 13:34Speaker 1: AI at all still or they're using it to create
- 13:38Speaker 1: a whole bunch of workslop.
- 13:40Speaker 3: So this is something that people are telling you because
- 13:42Speaker 3: you also asked not only have you received workslop, but
- 13:46Speaker 3: you're asking people fairly directly, are you generating AI work
- 13:52Speaker 3: that maybe is unhelpful and just passing it off like
- 13:54Speaker 3: it's your own work? And people are saying yes to this. Yeah.
- 13:57Speaker 1: What's incredible is that even with the self report or
- 14:00Speaker 1: biases that exist, people are still admitting to producing, you know,
- 14:05Speaker 1: just under twenty percent of their work it's WORKSLOP.
- 14:10Speaker 3: And I am going to assume that there are people
- 14:13Speaker 3: who are doing this and they're just not willing to
- 14:15Speaker 3: admit it.
- 14:16Speaker 2: I completely agree.
- 14:18Speaker 1: I will say that what we're starting to see is
- 14:20Speaker 1: the biggest predictor of creating WORKSLOP is actually being kind
- 14:26Speaker 1: of burned out. So people say that, like the reason
- 14:30Speaker 1: why they're generating it themselves as authors of WORKSLOP is
- 14:35Speaker 1: because these tasks that they're using AI to do are
- 14:39Speaker 1: one of many things that they're responsible for doing, so
- 14:43Speaker 1: they have a really full plate. You know, there have
- 14:45Speaker 1: been layoffs in the past few years, and so people
- 14:49Speaker 1: have increased span of control. They're having to do the
- 14:51Speaker 1: work of multiple people, and that makes it really tempting
- 14:55Speaker 1: to try to get a tool to do work on
- 14:57Speaker 1: your behalf. Also, I don't know if this feels true
- 15:01Speaker 1: for you, but a lot of what we hear from
- 15:04Speaker 1: our members and organizations is that, like everything feels really urgent,
- 15:08Speaker 1: and that's another predictor of producing work slow. It's just
- 15:12Speaker 1: the pressure to perform right now is really high, and
- 15:17Speaker 1: people are really depleted and haven't recovered from COVID.
- 15:21Speaker 2: It's really tempting.
- 15:22Speaker 1: When you have a very powerful tool to rely on
- 15:26Speaker 1: it to help you out. It's just that people are
- 15:30Speaker 1: doing that in a really blind way.
- 15:32Speaker 3: I mean, I would go further than saying it's tempting.
- 15:35Speaker 3: I would say it's the logical decision.
- 15:37Speaker 1: Right.
- 15:38Speaker 3: If you were at a job that you're watching people
- 15:41Speaker 3: getting laid off, you think you might be next, and
- 15:45Speaker 3: you're doing multiple people's job and you're not getting paid anymore,
- 15:50Speaker 3: why not use AI to get some of that stuff
- 15:54Speaker 3: off of your plate. What we're talking about here is
- 15:57Speaker 3: we're talking about fundamentally, I think in empathy thing which
- 16:01Speaker 3: is workslop. Maybe just the sign that, Yo, this person's overworked,
- 16:06Speaker 3: that's what I should think, But my immediate knee jerk
- 16:10Speaker 3: reaction might be this person's lazy, that're a bad person,
- 16:12Speaker 3: they don't respect me, right, And how do you even
- 16:16Speaker 3: get around that?
- 16:17Speaker 1: It's such a compassionate reframe on the situation, and maybe
- 16:21Speaker 1: you just came up with a new intervention. We could
- 16:24Speaker 1: try to do an experiment on that of just you know,
- 16:26Speaker 1: when people, you know they're about to start judging people,
- 16:30Speaker 1: that maybe we could just introduce the idea that people
- 16:33Speaker 1: are producing workslop because they're overworked, because everything feels urgent
- 16:38Speaker 1: and important, and maybe just by telling people that it
- 16:41Speaker 1: can initiate a dialogue that's more about hey, can I
- 16:45Speaker 1: help you approach these tasks in a different way.
- 16:48Speaker 2: Can I take something off of your plate?
- 16:50Speaker 3: I love your optimism.
- 16:53Speaker 2: My collaborator is Canadian. I can't help it.
- 16:58Speaker 3: I mean, I mean real, just real, Because maybe the
- 17:01Speaker 3: workslop thing isn't necessarily a sign that everybody's using chatgabt
- 17:07Speaker 3: or the chatchabt is bad or the claud is bad.
- 17:10Speaker 3: Maybe that's not really it. Maybe it's more of a
- 17:12Speaker 3: deeper fundamental societal thing, which is that yo, people are overworked.
- 17:19Speaker 2: I think that's exactly right.
- 17:20Speaker 1: Like that has been the focus of our research for
- 17:23Speaker 1: the last year, is about this crisis that we're in.
- 17:27Speaker 1: We've seen this over the past ten years in our
- 17:29Speaker 1: data that people are decreasing in the amount of resilience
- 17:33Speaker 1: and agility and these foundations of well being they have.
- 17:37Speaker 1: We're an all time low. So people really lack motivation, optimism, agency,
- 17:42Speaker 1: and we're just hungry for compassion and we just need
- 17:45Speaker 1: someone to refuel us so that we can be motivated
- 17:49Speaker 1: and feel like we matter and our work matters.
- 17:52Speaker 3: Again, basically, the conditions that are causing the problem here,
- 18:00Speaker 3: like we are going to continue to create those conditions.
- 18:04Speaker 3: The product is the output, which is the product which
- 18:07Speaker 3: is the output with this sounds like a spiral to.
- 18:09Speaker 2: Me, a vicious cycle. I think there's a way out.
- 18:14Speaker 1: We are thinking a lot about what type of interventions
- 18:17Speaker 1: are possible to prevent work slog and there are a
- 18:21Speaker 1: few things that are possible.
- 18:23Speaker 3: We'll get into that after the break. So Sam Altman said,
- 18:38Speaker 3: I believe, paraphrasing him, that this year we may see
- 18:42Speaker 3: AI basically quote unquote join the workforce. You've studied how
- 18:48Speaker 3: actual employees feel about basically AI kind of joining the workforce.
- 18:53Speaker 3: Where are we in terms of that right now?
- 18:56Speaker 2: Actually, you may be surprised.
- 18:58Speaker 1: I think that people are really excited about AI joining
- 19:02Speaker 1: the workforce.
- 19:03Speaker 2: When we ask people about.
- 19:05Speaker 1: This idea of managing a hybrid workforce of humans and agents,
- 19:10Speaker 1: for the most part, they express positive emotions like excitement, optimism, confidence.
- 19:15Speaker 3: Well, when you say people here, are we talking managers?
- 19:17Speaker 3: Are we talking CEOs, bosses? Are we talking workers?
- 19:21Speaker 1: Okay, that's a great question. Primarily when we ask managers,
- 19:25Speaker 1: they're more excited than individual contributors about AI joining the workforce.
- 19:31Speaker 1: And what people are wanting is training, And so we
- 19:35Speaker 1: have done a lot of work trying to understand what
- 19:39Speaker 1: type of training people need in management skills to best
- 19:42Speaker 1: manage AI, and for the most part, relational skills training,
- 19:48Speaker 1: a much more human type of training providing context, listening,
- 19:54Speaker 1: empathy is far more effective than task based training, where
- 19:59Speaker 1: you're giving directions on how to prompt, how to specify style,
- 20:04Speaker 1: how to evaluate the AI's output accurately. There are all
- 20:08Speaker 1: these courses you can take right now in AI literacy.
- 20:11Speaker 1: They're all about these task based training, hard skills to
- 20:15Speaker 1: learn how to do this. We find then if you
- 20:18Speaker 1: teach people to be really good managers, they're going to
- 20:21Speaker 1: do a really good job managing the AI and have
- 20:24Speaker 1: better results.
- 20:27Speaker 3: So really a big part of reducing workslot falls on
- 20:30Speaker 3: the managers and the CEOs. They're the ones responsible for
- 20:33Speaker 3: creating an environment where people can be comfortable with how
- 20:36Speaker 3: AI is being used, or where people can be comfortable
- 20:39Speaker 3: speaking up if something's wrong.
- 20:42Speaker 1: One of the things that we're seeing is that having
- 20:44Speaker 1: an environment that's psychologically safe, where people have the ability
- 20:49Speaker 1: to ask questions and take risks in a way that
- 20:52Speaker 1: feels safe, so they don't have to be worried about
- 20:55Speaker 1: whether using the AI is demanded or appropriate or permissible,
- 21:00Speaker 1: you know. So, I think part of that is like,
- 21:03Speaker 1: can I try this out? Can I disclose that I've
- 21:05Speaker 1: used it here? Can you give me some feedback on
- 21:08Speaker 1: whether this quality is high enough? Also training those mindsets,
- 21:12Speaker 1: so the pilot mindset, teaching people to have these skills
- 21:15Speaker 1: to be augentic over their usage of AI and also optimistic,
- 21:20Speaker 1: so curious and confident and willing to explore different ways
- 21:24Speaker 1: to use it to be more creative as opposed to
- 21:27Speaker 1: feeling like you have to use it. And then I
- 21:29Speaker 1: often like to say that AI is a multiplayer tool,
- 21:33Speaker 1: multiplayer game. And if we remind people that there is
- 21:38Speaker 1: a human recipient on the other end of your work,
- 21:41Speaker 1: and your goal is to collaborate with another human, and
- 21:45Speaker 1: you think about the potential negative impact of producing thoughtless
- 21:49Speaker 1: work that's AI generated, then I think even that psycho
- 21:53Speaker 1: education and the salience of knowing that there are humans
- 21:57Speaker 1: on the other side of the tool can be really
- 22:00Speaker 1: helpful to having positive collaboration.
- 22:02Speaker 3: Again, I think the study in some ways is a
- 22:06Speaker 3: kind of a warning. And basically, right now I can
- 22:09Speaker 3: totally understand why somebody would generate work slap, and it
- 22:15Speaker 3: would be great if their bosses or their company would
- 22:20Speaker 3: recognize the root cause of why they're using chat GBT
- 22:25Speaker 3: to write code or write a presentation that would be amazing.
- 22:30Speaker 3: Let's be realistic, that probably won't happen. And so your
- 22:34Speaker 3: study finds forty percent of people have received work slap
- 22:38Speaker 3: in the last month. I see that going up. Let's
- 22:42Speaker 3: talk about a five year in the future period. What
- 22:45Speaker 3: does that look like.
- 22:46Speaker 1: Yeah, it's a really complicated moment that we're in right now,
- 22:51Speaker 1: and I think the pressure to do more with less
- 22:54Speaker 1: is increasing, and there are so many unknowns about the
- 22:59Speaker 1: future of humans and the way that we work together and.
- 23:03Speaker 2: The tools that we use.
- 23:04Speaker 1: And they don't mean to be pollyannish and thinking about,
- 23:08Speaker 1: you know, the way that we can help and the
- 23:10Speaker 1: way that we can train people. I do think there
- 23:12Speaker 1: are deeper issues to bring like politicians and policymakers in.
- 23:17Speaker 1: But I have to say that, like, we've been through
- 23:21Speaker 1: this before. So think back to the early two thousand
- 23:27Speaker 1: and tens when social media came around and people had
- 23:31Speaker 1: to people meeting organizations and the people within them had
- 23:35Speaker 1: to learn to communicate and collaborate in a new way.
- 23:38Speaker 2: You know, it reshaped the whole.
- 23:40Speaker 1: World of just as an example, marketing and advertising, it
- 23:44Speaker 1: was a totally enormous shift in the way that we
- 23:48Speaker 1: do work, the way that we relate to each other,
- 23:50Speaker 1: the way that we use tools to get our work done.
- 23:53Speaker 1: And I think that gives me hope that you know,
- 23:56Speaker 1: this is a really, really powerful tool. We can create
- 24:01Speaker 1: a ton of value. We may need to slow down
- 24:05Speaker 1: a little bit, but you do have to stay relevant
- 24:09Speaker 1: with the tools that we all have access to right now,
- 24:11Speaker 1: to test them out, to try them out, to figure
- 24:14Speaker 1: out if you know, in the same way that years
- 24:17Speaker 1: ago we learned code, maybe no longer as relevant, but
- 24:22Speaker 1: I think you have to figure out how to use
- 24:24Speaker 1: these tools.
- 24:26Speaker 3: I mean, yeah, that's kind of the wild thing, is
- 24:27Speaker 3: I mean, I remember when a journalist lost their job,
- 24:30Speaker 3: you know, somebody to get up on Twitter and say, hey,
- 24:32Speaker 3: learn to code.
- 24:33Speaker 1: Ha ha ha ha ha.
- 24:34Speaker 4: Right.
- 24:35Speaker 3: Ironically, some of those same people I think are probably
- 24:39Speaker 3: now looking for a gig along with the rest of us,
- 24:42Speaker 3: because entry level coding gigs, I mean Claude and chat
- 24:46Speaker 3: GBT kind of has that on lock. It may be
- 24:48Speaker 3: generating terrible code, but it may be quote unquote good enough,
- 24:52Speaker 3: or as you say, it may be workslop and then
- 24:55Speaker 3: somebody gots to come in and clean it up. But
- 24:57Speaker 3: the company hasn't quite figured that out yet.
- 25:00Speaker 1: Right, So maybe there's a new industry of people who
- 25:04Speaker 1: have strong critical thinking skills and the ability to evaluate
- 25:07Speaker 1: the quality of code and to detect these signals and
- 25:10Speaker 1: to point that out. You know, I could see that
- 25:12Speaker 1: being an unexpectedly useful ability. I noticed that, Like, my
- 25:17Speaker 1: kids are way better at detecting AI generated images whenever
- 25:21Speaker 1: they are those games.
- 25:22Speaker 2: You know, AI generated or not.
- 25:24Speaker 1: I get them wrong every time, and my kids know
- 25:26Speaker 1: every single time, Like it's so obvious.
- 25:29Speaker 2: So maybe there are.
- 25:30Speaker 1: Some cues that you know, younger demographics, people who have
- 25:34Speaker 1: more hourly jobs can develop to help them evaluate and
- 25:38Speaker 1: work on their critical thinking skills to help the rest
- 25:41Speaker 1: of us prevent it.
- 25:43Speaker 3: I mean, I'm starting to wonder if maybe the one
- 25:45Speaker 3: prediction that we could make is that there might be
- 25:48Speaker 3: some kind of new category of job created, which is
- 25:52Speaker 3: basically work slop manager, like somebody who can who can
- 25:56Speaker 3: catch the work slop and say, I know what that is,
- 25:58Speaker 3: let me fix that workslop Jim workslop janitor. You gotta
- 26:02Speaker 3: be kidding me, man, You're right, I think some people
- 26:05Speaker 3: are that right now.
- 26:06Speaker 2: I think they are for sure that it's.
- 26:09Speaker 3: A little bleak, but maybe it's necessary. Maybe this is
- 26:13Speaker 3: where we're headed. Maybe not everybody figures out a way
- 26:15Speaker 3: within their company to make it so the workslop isn't
- 26:18Speaker 3: being generated, and so just understanding that this stuff is
- 26:21Speaker 3: going to be generated, and then we got to have
- 26:23Speaker 3: somebody clean it up before it gets put out. You know,
- 26:29Speaker 3: Kate and I didn't realize this when we were talking,
- 26:31Speaker 3: but it turns out that the phrasing of janitors for
- 26:34Speaker 3: AI workslop is actually already a thing. I was just
- 26:37Speaker 3: searching and there is a site called code janitor that'll
- 26:40Speaker 3: charge you a couple thousand dollars to clean up the
- 26:42Speaker 3: AI generated code so that your app will stop crashing
- 26:45Speaker 3: and actually work. This is a niche job that I
- 26:48Speaker 3: think is probably going to spread out to other industries,
- 26:51Speaker 3: maybe even the industry that you work in, to which
- 26:54Speaker 3: you might say, well, wouldn't it just be better to
- 26:57Speaker 3: have hired somebody who's good at their job to do
- 26:59Speaker 3: it right there the first time? Yes, probably, but if
- 27:03Speaker 3: you're not the boss, that decision isn't always up to you.
- 27:09Speaker 3: Thank you for checking out another episode of kill Switch.
- 27:12Speaker 3: If you want to email us we're at kill Switch
- 27:14Speaker 3: at Kaleidoscope dot NYC, or we're also on Instagram at
- 27:18Speaker 3: kill switch pod. And if you like the show, which
- 27:21Speaker 3: hopefully you do because you're all the way at the end,
- 27:24Speaker 3: think about leaving us a review. It helps other people
- 27:26Speaker 3: find the show, which helps us keep doing our thing.
- 27:29Speaker 3: And once you've done that, did you know that kill
- 27:31Speaker 3: Switch is on YouTube? So if you like seeing stuff
- 27:34Speaker 3: as well as hearing it, you can have the link
- 27:36Speaker 3: for that. In the show notes, kill Switch is hosted
- 27:39Speaker 3: by Me Dexter Thomas. It's produced by Sena Ozaki, Darluck Potts,
- 27:43Speaker 3: and Julia Nutter. Alexanderveld also helps with production on this episode.
- 27:47Speaker 3: Our theme song is by me and Kyle Murdoch and
- 27:50Speaker 3: Kyle also mixed the show. From Kaleidoscope, our executive producers
- 27:54Speaker 3: are Ozma Lashin, Mangesh Hajikadur, and Kate Osborne. From iHeart
- 27:58Speaker 3: Our executive producers ur Katrina Norvile and Nikki E. Tor Oh.
- 28:03Speaker 3: And one last thing, I don't want people to get
- 28:05Speaker 3: the impression that workslop is just people turning bad work
- 28:08Speaker 3: into their bosses. The study shows that workslop also flows
- 28:12Speaker 3: down the organizational ladder too. Kate, give me an example.
- 28:16Speaker 1: There are some people who talk about receiving them from
- 28:20Speaker 1: their ceo.
- 28:21Speaker 2: Here's an example.
- 28:23Speaker 1: The CEO of our company sent over some notes for
- 28:25Speaker 1: fundraising strategy. It was just a whole lot of words
- 28:28Speaker 1: saying nothing Chat GPT puke he sent over would have
- 28:32Speaker 1: been so much more helpful as a paragraph written in
- 28:34Speaker 1: his own words. The style of writing, reliance on bullet points,
- 28:38Speaker 1: the fact that our CEO readily admits using GBT for everything.
- 28:42Speaker 1: He had just gotten back from vacation. He was asking
- 28:46Speaker 1: GPT for parenting advice. So it's just like.
- 28:48Speaker 2: Going on and on.
- 28:51Speaker 3: I'm sorry, Chat GBT for parenting advice is you know.
- 28:56Speaker 3: I'm gonna leave that alone because that's an entirely different episode.
- 28:59Speaker 3: I don't want to touch that right now. Wow, I really
- 29:03Speaker 3: hope that we do not need to make an episode
- 29:05Speaker 3: about people using chat GBT for parenting advice. Please do
- 29:09Speaker 3: not make us make an episode about that anyway.
- 29:12Speaker 4: Catch on the next one. Good Bye,