00:00This practice has been going on forever, right? People have been polishing their resumes for as long as there have
00:05been resumes. But what's interesting here is that, you know, after the introduction of chat GPT, we really saw a
00:12surge in words associated with AI. So LLM, AI, GPT, people were going back and adding these to jobs that
00:22they had left.
00:23And this is not sort of like fresh in their memories. This was like a medium of jobs they had
00:29left four years prior. So something really interesting happening there.
00:34So employers, I guess the folks that actually use LinkedIn to find employees, are they getting hip to this? Are
00:41they dealing with it, recognizing it?
00:43So I think there's a couple ways to think about this. One, this whole thing came up because researchers who
00:50study the labor market use LinkedIn now as data. It's a treasure trove for them.
00:55But this paper examines this idea of how much do these edits actually change our understanding of the labor market?
01:03So they found, for example, that if based on the edits, the time traveling that people have done, we would
01:10be overstating people's AI skills in 2022 by 30%.
01:15So it doesn't so much give us a great understanding of the labor market of the past as much as
01:22it does maybe an understanding of what employers are looking for today or what they're not looking for.
01:27Because that's what people are reacting to, right? It's like, oh, I think that employers want AI skills.
01:33So I'm going to make sure that it looks like I have those capabilities.
01:37And rewriting history in the process.
01:40To Paul's point, I wonder how employers are looking at this when we don't even know if human recruiters are
01:47looking at this or whether bots are looking at people's resumes and kind of filtering out and filtering in resumes
01:54that they think are appropriate for jobs that they want to have to feature AI experience.
02:00I think we are in this AI doom loop where we have people using, as we know, AI to write
02:08their resumes, write their LinkedIn profiles in order to apply for jobs that will be screened by AI to make
02:16sure that they fit the job descriptions that were also written by AI.
02:21So I don't know that we're I don't know what this is telling us.
02:24We're not getting a lot of really good information, but I do think that employers, we are seeing an increase
02:30in jobs that are posted that are related to AI, that are asking for human capabilities, right?
02:36Empathy, creativity.
02:38The problem is that I don't know how AI assesses that, right?
02:43So we're in a very difficult cycle right now.
02:46So, Beth, does LinkedIn assume any responsibility or liability for making sure stuff is actually accurate?
02:55Oh, Paul, they're just a platform.
02:56They're just a platform, right?
02:57They're just a platform.
02:58They're just a platform.
02:59And I should say, I'm not making any moral assessment about this, right?
03:02Because I do think, again, like people have been tailoring their jobs to whatever the job they're applying for.
03:11There's obviously different degrees of that.
03:13So it's and again, I don't think people I think it is hard to go back and say, hey, I
03:18had all these AI skills 10 years ago, right?
03:21But I do I do think that the practice itself is not is not new.
03:26But the way it's so public, I think, is very new.
03:30You're not just sending it to one employer.
03:32You're putting out out there for everybody to see.
03:35And the interesting part of this research study was that this is happening in particular to the tech industry.
03:42People in the tech field who are looking for jobs, where I guess it's more credible that they might have
03:48been doing something with AI and large language models four years ago or five years ago than, say, if Paul
03:53or I put it in our resumes.
03:55Totally, totally.
03:56Yeah, I was like one, I think for tech.
03:58So it was 20 percent for the overall popular, you know, LinkedIn user population in the US.
04:03But it was one in three for tech workers.
04:06But, you know, I think that it can tell us it does tell us information about the labor market, even
04:12if it's not so much a historical artifact.
04:15Right. We could say, you know, if we saw a surge in people update.
04:20So people do tend to update their their LinkedIn profiles before changing jobs.
04:24There there's a correlation, you know, there's there's a high connection there.
04:28So what they found what, you know, I think one way to think about this is if we see a
04:33surge in people updating their LinkedIn profiles, does this foretell the next great resignation?
04:40Or could you tell something about a company's health if you were looking at a particular company and saw a
04:46lot of employees updating their LinkedIn profiles?
04:48Does this mean more?
Comments