Ep 39: Bridging the Skills Gap and Empowering Women in AI with Isabelle Bichler

[00:00:00] Welcome to Elevate Your AIQ, a podcast focused on the AI-powered yet human-centric future of work.

[00:00:05] Are you and your organization prepared? If not, let's get there together. The show is open to sponsorships from forward-thinking brands who are fellow advocates for responsible AI literacy and AI skills development to help ensure no individuals or organizations are left behind. I also facilitate expert panels, interviews, and offer advisory services to help shape your responsible AI journey. Go to ElevateYourAIQ.com to find out more.

[00:00:28] Hi everyone, welcome back to Elevate Your AIQ. Today for this episode I'm excited to have Isabelle Bichler, co-founder and COO of Retrain.ai. I first met Isabelle in early 2023 at the Responsible HR Summit in New York City that Retrain.ai hosted, and this event was one of the experiences that convinced me that Responsible AI was a critical focus area, not just for me, but for all of us trying to mitigate bias and promote fairness and ethics in our decisions.

[00:01:04] Isabelle and the Retrain.ai team are tackling one of the biggest challenges in the workforce today, which is the skills gap.

[00:01:09] In our conversation, we dive into how AI can bridge this gap responsibly, enhance human capabilities, and create more opportunities for continuous learning and growth.

[00:01:19] Isabelle also shares her thoughts on the importance of data-driven decision-making in HR, the cultural shifts needed for AI adoption, and the role of women in shaping the future of AI.

[00:01:29] This is an insightful discussion packed with ideas on how we can innovate responsibly while empowering the workforce of tomorrow. Thanks for listening.

[00:01:35] Hello, everyone. Welcome to another episode of Elevate Your AIQ. I'm your host, Bob Pulver. With me today, I have the good fortune to speak with Isabelle Bichler.

[00:01:47] She is the co-founder and COO of Retrain.ai. How are you doing today, Isabelle?

[00:01:53] Great, and great to be here. Thank you for having me.

[00:01:56] Absolutely. Thanks for joining me.

[00:01:57] One of the first things that, well, we'll get into this as the conversation goes along, but certainly I've always appreciated your advocacy for responsible AI in HR and beyond, because these are decisions being made about people and their livelihoods.

[00:02:14] And it's just important we treat that with most sensitivity.

[00:02:18] So thank you for that.

[00:02:19] Just to kick things off, why don't you just give us an introduction about your background.

[00:02:25] You've been working in and out of AI and technology for a while.

[00:02:29] Yes. Well, in my background, I've always been in the intersection of people at tech.

[00:02:34] The first company I established many years ago when I was 24, also in HR tech, bringing together a contingent workforce with job opportunities.

[00:02:43] I sold it three years after to a big staffing company.

[00:02:46] And since then, I was actually on the other side of the table as an investor, as a wealth manager, investing in tech, early stage startups, usually, but really different spaces across the board, you know, cyber, prop tech, fintech, and so forth.

[00:03:02] And I did this for 12 years.

[00:03:05] And since COVID hit 2020, together with my co-founders, Dr. Shane Gavid and Abby Seymour, we established Retrain AI to solve what we call one of the biggest problems that exists is the skills gap.

[00:03:18] And we established Retrain, and we're going to talk more about how we're helping a lot of people.

[00:03:25] Yeah, no doubt.

[00:03:26] Thank you for that.

[00:03:27] Really, really interesting.

[00:03:28] So you've got a lot of different sort of lenses that you can look to the market, not just the client challenges, but really the broader base and where people are investing and hopefully investing in the right solutions.

[00:03:38] I have, you know, the legal compliance side, and I did an MBA, and I also graduated from NYU.

[00:03:45] I did the Master of Science in Risk Management, which one of the research I've done there was about Responsible AI, about the risks of different technologies, AI, and mitigations, strategies.

[00:03:59] We're also part of the World Economic Forum, part of the, it's called the Innovator Program, basically discussing and standardizing the whole domain of Responsible AI, because there's many different angles to that.

[00:04:12] Sure.

[00:04:13] And also part of the Responsible AI Forum discussion.

[00:04:17] So it's really, it's an interesting time in terms of AI, and in terms also of ethical, responsible AI.

[00:04:23] Wow, that's a lot.

[00:04:25] So one of the things that I thought we could talk about is, I think because of that broad lens that you have, that you've got an eye to, you know, the, you know, the legal, you know, regulatory frameworks that are, that are coming, being somewhat multicultural.

[00:04:42] I'm imagining you're traveling all over the world, and you're hearing, you know, different perspectives from clients and prospects, community members about, you know, how this is evolving.

[00:04:52] What are you hearing in terms of either similarities or differences from maybe different countries, different cultures?

[00:05:00] I think the one thing that is always there is the problem of not having skilled talent.

[00:05:08] And this is a concern, not just for companies, it's also for governments, right?

[00:05:14] To lead a country when in the era of very, very fast changes in technologies, geopolitical, you know, conflicts, you can say, now that are happening all over.

[00:05:27] However, you definitely want to be on top of it and make sure that your citizens, your workers are productive and are efficient and have the right skills.

[00:05:39] So the, it was always a problem, right?

[00:05:42] But it's accelerated and it's now something that, and COVID was part of it also one of the factors that really accentuated that problem.

[00:05:49] So, so I think it's on top of mind of everybody.

[00:05:52] And of course, all the stats of 77% of CEOs are thinking are, you know, that's a top concern, having skilled talent.

[00:06:00] And of course, it's cascading into talent's needs and skills, right?

[00:06:05] So, and that's why we're talking, we're talking about skills, intelligence, how AI is helping HRs hire faster, retain longer, and so forth.

[00:06:14] So that's what Recreate AI are doing.

[00:06:16] Yeah, so the skills piece, I mean, that's central to, to everything, right?

[00:06:21] I mean, you've got to get a better handle on not just where your current gaps might be, but does the trajectory of, of AI, you know, capabilities or competencies?

[00:06:34] I'm not sure how curious to get your take on how you think about that.

[00:06:37] But, you know, as that comes in, it's not just like some kind of annual, you know, exercise, right?

[00:06:42] You've got to constantly be looking at who can we upskill, who can we reskill?

[00:06:46] Do we have to hire net new?

[00:06:47] Do we use an agent, digital, you know, worker of some kind to do that?

[00:06:52] It's becoming more complicated.

[00:06:53] And that's exactly what led us to really establish Retrain.

[00:06:58] The inception point was looking at what's going on in the market, seeing those fast changes, understanding that you can say 44%, almost half of the individual worker skills are going to be needing a significant upskilling and reskilling until 2027.

[00:07:15] In order for us to be employable, let us to understand that there's a very big problem.

[00:07:21] But specifically in HR, in HR world, in enterprise HR world, there's a lot of systems.

[00:07:27] It's pretty distributed and those systems are siloed.

[00:07:31] There's a lot of data maybe in the systems, but they're not utilized and you don't know enough about your talent.

[00:07:37] And that's something that for me, you know, I'm always as a financial person, right?

[00:07:42] I'm doing investment.

[00:07:43] I'm looking at the stock market.

[00:07:44] I'm looking at, you know, the market and different assets and seeing the data in real time, understanding trends.

[00:07:52] And this is something that is not available for HRs.

[00:07:54] So they have a lot of data.

[00:07:56] They're sitting on tons of data, whether it's in the organization or just, you know, publicly available data.

[00:08:01] And it wasn't collected, analyzed in any way.

[00:08:08] So that was the effort that we've done.

[00:08:10] And that's how we established HR and AI in mind of bringing data for insightful predictions and recommendations for HRs to make decisions.

[00:08:21] I've had a few conversations with folks in like the talent intelligence community or the people analytics community.

[00:08:27] It seems so logical to me and definitely logical to you because you started a company to focus on this area.

[00:08:35] But part of it seems like it's so logical to bring this data together.

[00:08:39] And then it's really about do you have the right culture and maybe organizational structure to make sure that you're recognizing these benefits and overcoming whatever challenges and risks may exist to get to this end state where you do have that much better and, you know, consistent, like perpetual visibility to all of that and all that insight.

[00:09:05] Right.

[00:09:05] Yeah.

[00:09:06] And for sure, it's a cultural thing.

[00:09:08] And it's also coming from an eccentric need to see some specific industries where they have to do something about it.

[00:09:15] And if they want to stay here, viable.

[00:09:17] Right.

[00:09:19] Healthcare, manufacturing, those changes there are critical.

[00:09:22] So those kind of factors brought those early adopters to say, hey, let's find a solution and collaborate with startups that are bringing that innovation and disruption and try and see and explore how we can solve those big problems.

[00:09:39] Yeah, for sure.

[00:09:40] So it seems like when your clients implement this solution, I mean, you've got to be incredibly eye-opening for them, right?

[00:09:49] It's like, why did it take us so long to even, you know, do this or whatever?

[00:09:54] I mean, any particular call outs from some of the, you know, client feedback?

[00:10:00] You're talking about stories and use cases of implementation.

[00:10:04] Yeah.

[00:10:04] So I was just curious if there were some that were just like over the top, like I can't believe we even questioned this investment, right?

[00:10:12] Oh, for sure.

[00:10:13] So I think what we've learned is that HRs need to have very specific KPIs for them to see the value, right?

[00:10:22] It was discussed in the work, kind of a broader, high level in the beginning, non-tangible outcomes.

[00:10:28] And we bring it to the, depending on the use case, of course, if it's for hiring.

[00:10:32] So what do we want to achieve?

[00:10:34] Is it hiring better people in terms of the qualified talent that they need for specific criteria?

[00:10:42] Is it the faster time to hire, bringing efficiency?

[00:10:46] Is it both and so forth, diversity into the pipeline and so forth?

[00:10:50] Definitely what we do, and maybe I'll say a quick summary about Retrain AI.

[00:10:55] So Retrain AI established 2020.

[00:10:58] We basically help HRs hire faster, retain longer, develop the workforce much better.

[00:11:03] And we have, we're leveraging Responsible AI for that.

[00:11:07] We have three solutions.

[00:11:09] One solution is really building a ground level, we call it, a skills architecture, a skills taxonomy.

[00:11:15] We generate it automatically.

[00:11:17] And of course, the organization can calibrate and adjust and configure it to their needs.

[00:11:22] And we have a talent acquisition solution, basically providing a search and a match, ranking of the pipeline of the candidates and explainability.

[00:11:33] There's explainable AI, explaining why that person is recommended to the position, the open position.

[00:11:38] And a talent management solution, basically creating a marketplace within the organization, connecting employees to job opportunities,

[00:11:46] whether it's project, an open position, learning, and so forth.

[00:11:49] And all that provides insights, right, to HR.

[00:11:52] So on each of those models, we have very specific KPIs to show value.

[00:11:57] So for hiring, for example, the time to hire, we work with a very large company.

[00:12:03] And usually we would come and integrate to the system of record.

[00:12:06] So we ingest the data from the system of record and provide the insights about what we think about that person.

[00:12:14] Is it a good recommendation?

[00:12:16] Should we move forward with that person?

[00:12:17] Yes, no, and why?

[00:12:18] Yeah.

[00:12:19] Tremendous results in terms of you've now lowered.

[00:12:22] If 40% of the time of a recruiter is spent on search, now we've lowered it by 50%, 55%, whatever it is.

[00:12:30] So that's significant efficiency we bring into that.

[00:12:33] So that's what we're tracking constantly.

[00:12:35] Awesome.

[00:12:35] So I didn't realize you had some of the capabilities that people might look to like a talent, internal talent marketplace for.

[00:12:42] I mean, that would be the purpose of or the benefit of pulling all of that data together, like you said, to do the skills development, to look for a mentor, to understand, you know, maybe some career paths and opportunities that you hadn't thought of because it's maybe it's a lateral or maybe it's a diagonal move.

[00:13:02] Right.

[00:13:04] Right.

[00:13:33] Right.

[00:13:34] I was curious to get your take on that.

[00:13:36] So people call it power skills or, you know, there's a lot of names to that.

[00:13:41] And I think it's about the ability to adjust and learning, continuous learning.

[00:13:47] And so it can be AI for sure.

[00:13:49] We need to have that ability to be flexible, to be adjustable, to be adaptable, right?

[00:13:54] And to learn fast.

[00:13:56] And we see also, by the way, that women, there's only a third of them that is using, if you compare it to male, those kind of tools, GPT, whatever it is, LLMs.

[00:14:09] And that's a problem.

[00:14:11] So women actually need to augment that ability, not be afraid of that.

[00:14:16] But I do think I'm very optimistic about the usage of AI.

[00:14:20] I think people are taking this as an opportunity now to upskill and quickly learn.

[00:14:25] And you can see the benefits, although some of them are still errors, hallucinations in the LLMs we're using.

[00:14:33] But still you see the benefits.

[00:14:35] And so it incentivizes everybody to use more and more and be more efficient.

[00:14:39] So those skills, I think, are super important.

[00:14:42] I was just working with a group.

[00:14:44] I signed myself up.

[00:14:46] I got to take my own advice.

[00:14:47] I signed myself for an agentic workflow sort of boot camp.

[00:14:51] I was curious, I guess, how you see maybe the agentic workflow stuff evolving and how that may alter, not necessarily the in-demand.

[00:15:02] Hi there.

[00:15:03] I'm Peter Zollman.

[00:15:04] I'm a co-host of the Inside Job Boards and Recruitment Marketplaces podcast.

[00:15:10] And I'm Steven Rothberg.

[00:15:11] And I guess that makes me the other co-host.

[00:15:12] Every other week, we're joined by guests from the world's leading job sites.

[00:15:16] Together, we analyze news about general niche and aggregator job board and recruitment marketplaces sites.

[00:15:23] Make sure you sign up and subscribe today.

[00:15:27] Skills.

[00:15:28] I mean, it could add to some skills that would be valuable if you're going to implement something like that.

[00:15:32] But I guess I wonder if it alters the timeline for some of the projections.

[00:15:39] I think it would be much faster than what we're thinking.

[00:15:42] I think that's the difference between all the technologies that we've seen until now.

[00:15:46] The adoption was fast, but this one is going to be much faster.

[00:15:49] I think it's the tip of the ice, really.

[00:15:53] And I'm very excited about it.

[00:15:55] I think it's...

[00:15:57] And that's why, again, we've gone back to the skills we need.

[00:16:00] So continuous learning, fast learning, and be open to that innovation that is happening.

[00:16:06] Not be fearful of it.

[00:16:07] It's not going to replace you.

[00:16:08] It's going to enhance you.

[00:16:09] Of course, it's going to replace a few things that you would have done otherwise.

[00:16:12] But it's maybe better, right?

[00:16:15] Because you don't want to do that manual work.

[00:16:17] You don't want to do that tedious work.

[00:16:22] You don't want to do that.

[00:16:23] You don't want to do that.

[00:16:23] You don't want to do that.

[00:16:23] You don't want to do that.

[00:16:24] You don't want to do that.

[00:16:24] Why not?

[00:16:26] Who needs to do it, really?

[00:16:27] I find myself in an interesting spot.

[00:16:30] I'm curious to get your perspective on this as well.

[00:16:33] But I spent most of my career, and I continue to spend a lot of my time thinking about innovation.

[00:16:41] Not just innovation, but innovation management.

[00:16:43] How do you guide the best ideas forward and build teams around them and things like that.

[00:16:48] But more recently, with my responsible AI hat on, you've got to recognize that there's a balance.

[00:16:57] And so I do wonder, and I guess I'm cautiously optimistic that this will come to pass.

[00:17:03] But I just worry that people, when anyone can build something and they know they can start generating revenue from it, or is the very concept of an ease of setting up an agentic workflow going to trigger a whole other wave of, tidal wave perhaps, of just like shiny objects?

[00:17:26] Well, at the end of the day, it's the data that you're using in the algorithms.

[00:17:31] If you're pulling from data that is erroneous, you're going to get results that are not really good.

[00:17:39] But I think we're getting better at this.

[00:17:40] So you see GPT, for example, you still see hallucination.

[00:17:44] You will see.

[00:17:45] It's not trained on a specific context.

[00:17:47] So there are now very specialized LLMs, right?

[00:17:51] Or you see vendors taking it to the next level of making sure it's responsible.

[00:17:57] It has less errors.

[00:18:00] It's not biased.

[00:18:01] It's not causing any security or privacy breach, right?

[00:18:05] It's kind of under a specific guardrail.

[00:18:08] So, you know, you cannot stop the innovation.

[00:18:11] And it can coexist with regulation and some limitations.

[00:18:15] That's okay.

[00:18:15] And we just need to make sure we're using it correctly because not all tools are equal.

[00:18:22] So you want to make sure you're using the right tools for the right purpose.

[00:18:25] I guess one of the things I think about is instilling in founders or aspiring founders some of the things that you established from the beginning.

[00:18:35] So in a way, Retrain is one of those companies who are a good model to follow in the sense of being responsible by design

[00:18:43] and thinking about, you know, the ethics and the fairness of these decisions, the transparency of, you know, its output and things like that from the beginning.

[00:18:50] And I just, I hope, like I said, I'm cautiously optimistic that people follow that.

[00:18:55] I just, I'm bracing myself for some nonsense.

[00:18:59] Well, I think we started when it was really hard to talk about AI.

[00:19:04] People were not understanding yet in 2020, 2021.

[00:19:07] I was saying the word taxonomy.

[00:19:09] I was saying the word NRP.

[00:19:11] People were not understanding what I'm saying.

[00:19:13] Now I have people that understand very well and are looking for very specific and they have very intentional challenges that they want to solve with specific results and goals.

[00:19:24] So I'm very happy about that maturity level.

[00:19:27] And definitely when I was talking about responsible AI at the time, people were totally not getting it.

[00:19:33] And my co-founder was saying, they don't talk about it because people don't get it.

[00:19:38] And I said, but this is something critical and we're going to see that later on, it's going to be discussed.

[00:19:43] And now it's, you know, when everybody's talking about mine as well, because when you want to implement a technology,

[00:19:48] you want to make sure you assess it for the specific criteria of ethical responsible AI.

[00:19:53] So I'm happy about that maturity and adoption level now of all those firms and understanding what it means and how to more and more how to implement because it's still, it needs a lot of still understanding.

[00:20:06] Yeah.

[00:20:07] I think that I should not have been surprised that that was slow to enter people's consciousness and vocabulary.

[00:20:16] I mean, we weren't talking about this at all when I was at IBM and we were watching, you know, IBM Watson come out of the labs.

[00:20:23] That sort of predictive AI, analytical AI was different, right?

[00:20:28] Because the average person wasn't interacting with it and wasn't taking its output and running with it.

[00:20:34] I mean, yeah, we've had Google search results for quite a while and those were always, you know, suspect for certain, you know, inquiries.

[00:20:41] But I just feel like, yeah, it was definitely generative AI that really made people start thinking about, do I trust these things?

[00:20:49] And so, like you said, there's a maturity that needs to occur.

[00:20:54] And some people, luckily, many vendors are starting to catch on that this is what it means to be a trusted partner in the age of AI is to be responsible by design.

[00:21:02] Yeah, for sure.

[00:21:04] I was curious about the, just with your background with doing like contingent hiring.

[00:21:10] One of the things that I've talked to folks about is assessing, not just assessing skills, skills gaps, et cetera, but the skills intelligence that an organization may have.

[00:21:20] If you're using contingent labor through like a vendor management platform or something like that, if you were to connect, you know, retrain or just think, just thinking more broadly,

[00:21:29] if an organization is trying to look at, get a 360 degree view of all the available talent.

[00:21:36] How do you see like contingent, you know, playing into that?

[00:21:40] Is that like, you know?

[00:21:41] Well, that's a big piece, but we're not covering that piece.

[00:21:44] We're covering the ones that you have in your pipeline currently.

[00:21:48] So active pipeline, you have the passive ones.

[00:21:51] Actually, let's be very, very specific about past candidates, people that applied, let's say, two years ago.

[00:21:57] So we call it the omni-channel, meaning that we're bringing all those pipelines in one mainstream channel.

[00:22:04] And we also look at the internal employees.

[00:22:09] So you have also people with an organization of 80,000 employees, right?

[00:22:12] There's a lot of people that could be actually transitioning.

[00:22:15] Instead of leaving you to someone else, they could actually either upscale and move or just move.

[00:22:21] Because they're a good fit and you don't know about them.

[00:22:24] And they don't know about that opportunity that is there, available.

[00:22:29] And there's another stream, the stream of the passive, the ones that it's outsourced.

[00:22:33] We have a very large talent network that has more than 500 million resumes that we can tap into.

[00:22:40] So bringing all those four different pipelines into one omni-channel.

[00:22:45] So when a recruiter opens a position, opens a requisition, we understand the different criteria that we're looking for.

[00:22:53] And then we compare it to all those different candidates from all those different streams into one channel.

[00:23:00] So it's very efficient for the recruiter to pull everybody all together and see how they rank against that specific open requisition.

[00:23:06] So contingent workforce is definitely something that we have in mind.

[00:23:10] The VMS that exists would be a great source of data.

[00:23:14] Also external as well.

[00:23:16] So there's a lot of different networks of contingent workers out there.

[00:23:21] And the gig economy, you know, it's growing.

[00:23:23] It's definitely a very good market to go into.

[00:23:27] But we're focusing, we're very focused on what we're doing right now to make it better and better and better.

[00:23:32] And then we're going to add more pieces to it, such as the contingent workforce.

[00:23:37] But I definitely, many years ago, I was thinking this is a great market to go into.

[00:23:42] And that's why we see also the fiber of the world, all these different platforms.

[00:23:48] Definitely great success there.

[00:23:50] Yeah, I think that's certainly something that I pay attention to just because I think about,

[00:23:56] like, if you identify those skills gaps, being able to look, you know, anywhere.

[00:24:00] And, you know, if most, if a lot of the good talent prefers to be, you'll still find them on,

[00:24:04] you know, probably LinkedIn and, you know, other places.

[00:24:08] It's not like they're only, they only exist on the fibers and Upworks of the world.

[00:24:12] But certainly there's a huge amount of talent that chooses to work that way.

[00:24:17] And I was curious, just in terms of, you know, as you pay attention to everything that's going on,

[00:24:24] as you think about it, whether you're wearing your investor hat on or your lawyer hat on,

[00:24:28] as you think about some of the solutions that you're playing with,

[00:24:33] anything interesting that you have been playing with for yourself or for Retrain or just in,

[00:24:39] it could be in your personal life.

[00:24:40] I'm just curious if there's anything that piques your interest.

[00:24:42] Well, out of the HR tech, there's a lot of things that are making me very curious about what's going on.

[00:24:48] And healthcare, I think it's a huge, huge opportunity there.

[00:24:51] The way it works right now is thoroughly broken, but that's a sign.

[00:24:56] In HR tech, I think we're really the forefront of it.

[00:24:59] We're seeing bringing that and the vendors out there are also the ones I usually compete with are great.

[00:25:07] It's a great competition to be in.

[00:25:09] And everybody's trying to help HRs on that, you know, journey to find people, to retain people, to upskill people.

[00:25:17] We're all doing the best we can.

[00:25:19] And the models are improving.

[00:25:21] The generations, you know, are growing.

[00:25:24] So we're bringing really the most advanced algorithms there.

[00:25:28] And there are some vendors that need to restack.

[00:25:30] What makes me excited, it's not actually the shiny objects, not agents or chatbots.

[00:25:35] It's really the data and the results that we're seeing.

[00:25:38] If you want to match a person to a position, so the accuracy, the precision, the recall, all those metrics have to be, you know, high.

[00:25:46] You want to continuously improve on that.

[00:25:49] You want to have explainable AI and reduce biases.

[00:25:53] Constantly show recruiters why they could actually look for other candidates based on skills, based on different potential attributes.

[00:26:02] So that's, for me, exciting.

[00:26:04] So I'm not going after the, again, next shiny objects.

[00:26:08] There's always going to be more.

[00:26:10] I'm really looking into the data and the algorithms layer.

[00:26:13] That makes total sense.

[00:26:14] I think these hallucinations, we know, you know, the big vendors are working on that big, I mean, foundational model kind of vendors.

[00:26:22] I don't mean, you know, domain vendors.

[00:26:24] But I think there's definitely a lot of improvement there.

[00:26:28] And especially in talent acquisition, it just seems like now that you've got potentially like 10x the applications, right?

[00:26:37] And their AI generated SMAs and CVs and everything's starting to look the same, both in the inputs and the way it's evaluating.

[00:26:46] It's like, what are we doing here?

[00:26:48] Yeah, it's a play also of the data that you have and the sources that you're fusing altogether.

[00:26:55] 360 picture about the person and enrich the data.

[00:26:59] So we see we work with organization and it could be pretty challenging just to work with a large organization in terms, not just because, you know, they have a lot of stakeholders.

[00:27:08] And there's a lot of people that are going to make the decision and you need to consult ways and you need to bring everybody into the room and so forth.

[00:27:14] It's not just that.

[00:27:16] It's because the data is lacking, right?

[00:27:20] So even if you have an HCM, the data you have there about people is not enough or it's out of date or it's actually wrong.

[00:27:31] Have you ever been to a webinar where the topic was great, but there wasn't enough time to ask questions or have a dialogue to learn more?

[00:27:37] Well, welcome to HR and Payroll 2.0, the podcast where those post-webinar questions become episodes.

[00:27:43] We feature HR practitioners, leaders, and founders of HR, payroll, and workplace innovation and transformation sharing their insights and lessons learned from the trenches.

[00:27:52] We dig in to share the knowledge and tips that can help modern HR and payroll leaders navigate the challenges and opportunities ahead.

[00:27:58] So join us for highly authentic, unscripted conversations and let's learn together.

[00:28:03] It's just non-existent.

[00:28:05] So we need to enrich that data.

[00:28:07] So at the end of the day, we're a data company, right?

[00:28:09] So we help and we say, okay, don't worry.

[00:28:12] The hazard listing is on us.

[00:28:14] If you don't have the data about role descriptions, right?

[00:28:17] You don't.

[00:28:17] We have what we call our generic number, which is in the context of the industry.

[00:28:22] So we have all the different role descriptions for finance or for healthcare or for retail or for luxury goods, whatever it is, for semiconductors.

[00:28:31] We do have it.

[00:28:33] And then you can start from that.

[00:28:35] So the call start problem is solved there because you have a layer, a base layer to start with, to calibrate and configure, which we also develop tools to make it seamless.

[00:28:46] Because at the beginning, it was hard.

[00:28:48] You tell HRs, well, you need to now work and adjust it to your organization, unique value proposition.

[00:28:53] We don't want HRs to work for us.

[00:28:55] We want the technology to work for them.

[00:28:57] So we put a lot of efforts into having now calibration tools and it's improving constantly with the customers that we work with to make sure that we don't spend a lot of time on this.

[00:29:07] And they have that start, a seamless start.

[00:29:11] So the deployment, the data, the fact that you need to integrate into an ecosystem, right?

[00:29:17] So all that is definitely challenging, but it's part of what we're doing constantly to improve.

[00:29:25] That's awesome.

[00:29:26] People don't necessarily understand.

[00:29:28] I've been asked by some C-suite folks, like, is data governance still important?

[00:29:35] Is data quality still important?

[00:29:36] I'm like, okay, that's like the lifeblood of all of this, right?

[00:29:40] So the foundational capabilities, anyone who is up the sort of data and analytics sort of maturity curve is already much better positioned to be able to deploy AI in a responsible way and to get, you know, to mitigate, you know, hallucinations and have more trustworthy output right from the start.

[00:30:03] Yeah.

[00:30:04] So when we think about AI literacy, which I know, you know, you're a huge advocate for, and we think about what organizations need to do.

[00:30:14] I mean, as you go and talk to clients and prospects, and I know you do other, you know, sort of events and community types of things.

[00:30:22] How do you think about, you know, sort of upskilling and AI readiness overall for organizations?

[00:30:29] Well, I think now that if I had to draw it, I would say that, you know, the Gardner-Hive cycle is kind of a good visual for that.

[00:30:39] But to really show that the technology, the user expectations have in the maturity cycle.

[00:30:47] You know, we were seeing that people were not yet matured in terms of understanding AI, understanding how to deploy AI.

[00:30:55] Even the SEO skills-based organization concept was still very nascent.

[00:31:00] So you could say the spark of innovation, people started to get very excited about that.

[00:31:07] Again, if we talk about numbers, there's not a lot of them starting to do it in the early days, like three years ago.

[00:31:13] And still today, I think it's in their 30% deployment, really.

[00:31:18] Again, there's a lot of different stats, so I'm not sure what is the right one.

[00:31:22] But I tend to agree that it's not that much yet that have adopted, but they're now very into adoption.

[00:31:30] They want to start adopting.

[00:31:31] They understand the problem well.

[00:31:33] And in the maturity cycle, now after facing those, maybe the dissolution even happened after a few, one, two years and say,

[00:31:43] OK, yeah, it's not going to solve everything, OK?

[00:31:45] And it's not going to save humanity yet.

[00:31:48] It's the beginning.

[00:31:49] And, well, now that we understand what reality is, now we have realistic expectations.

[00:31:55] Now we can really, with those realistic expectations, have a deployment with realistic outcomes.

[00:32:01] But the beautiful thing is that there's another visual of another graph, linear graph, that goes like this,

[00:32:09] which is the performance of the technology.

[00:32:10] And even if you had that expectations and then the dissolution, and then again, now kind of realistic outcomes,

[00:32:17] now that performance is actually getting better.

[00:32:19] So I think now we're really meeting and we're actually better than the expectations.

[00:32:25] So, and the maturity is there.

[00:32:28] So now people want to adopt.

[00:32:29] They have great questions.

[00:32:31] I love to talk to HRs to learn from them and see how I can help and improve constantly.

[00:32:37] So that's kind of what I'm seeing in the market.

[00:32:40] In terms of literacy, I think I still, we still see that there's fear and still people, again,

[00:32:48] I talked about the gender, you can say gap there.

[00:32:51] And it's only a third of women are really using AI at work or at home compared to men.

[00:32:57] But it's an opportunity for us, you know, because if that's the problem, so let's have skill.

[00:33:02] I read a very interesting article.

[00:33:05] I was tagged on, I can send you the reference after, but it said good girls are good, the good girl syndrome or something like that using AI.

[00:33:16] And so women tend to think, and again, I'm not saying I'm specifically 100% agreeing with this, but I think it's an interesting aspect.

[00:33:24] Women tend to be, you know, want to be perfect.

[00:33:28] When you say something, you want to make sure it's accurate.

[00:33:30] That's kind of a women tendency versus men that they just throw it into the air and say, well, I'm not sure, but I'm saying it loud enough.

[00:33:37] So I guess I'm confident enough and it makes everybody feel comfortable too.

[00:33:41] But so women have that tendency to be perfect and not cheat.

[00:33:47] You don't ever to cheat on anything.

[00:33:49] So AI can make you cheat, right?

[00:33:51] Because you're cheating the system.

[00:33:52] You're summarizing where something else is doing it instead of you.

[00:33:56] So that could be one of those reasons for women not to use AI to feel that they're cheating.

[00:34:03] Specifically, I think it's actually augmenting what we can do and have, you know, more time to do the things that are meaningful.

[00:34:09] So good girl using AI could be part of that kind of explanation.

[00:34:16] Not sure it's true, but definitely interesting to read and try to understand what is it.

[00:34:21] But I think it's an opportunity in 70% or even more of HRs are women.

[00:34:25] So definitely we need to upscale continuously and then seize that opportunity.

[00:34:30] I got to check that out.

[00:34:32] It honestly doesn't surprise me, but I think it hits on one of the points that I always make, which is when we talk about like appropriate use of AI, like don't outsource your own critical thinking.

[00:34:45] And it sounds like men have a tendency to do that too often.

[00:34:50] But yeah, I mean, when you think about, of course, getting into the ethics of, you know, how and when to use it, you know, I've talked to people in the education system about this.

[00:35:01] I know it comes up a lot.

[00:35:03] It comes up with college admissions.

[00:35:04] It comes up in whether you're in undergrad, grad school, or even, you know, earlier than that in high school.

[00:35:10] So these are things that I think are important to think about.

[00:35:14] And then if we can get young women, you know, engaged even sooner when they're teenagers, then, you know, hopefully we can fix that stat.

[00:35:24] I totally agree.

[00:35:26] And I actually am a mentor in a few initiatives.

[00:35:29] There's an organization called Winistream.

[00:35:32] It's actually specifically in Israel, but it helps youth in the peripheral areas, geography and economic areas.

[00:35:40] To take, you know, kids from the age of 13 to 17 through high school to a program where they build products.

[00:35:48] And they actually launch it until the end of the, when they graduate.

[00:35:53] So bringing them into the center of the Israel tech, which is amazing, right?

[00:35:59] Absolutely.

[00:36:00] Gaining those entrepreneurial skills.

[00:36:02] So I was mentoring a judge, very, you know, involved in this organization and also mentoring youth here in New York, New Jersey.

[00:36:09] We, again, kids that are 13 to 17 building products over the year.

[00:36:15] It was great.

[00:36:16] And usually in purposes for helping other kids.

[00:36:20] So, and it's, it's that literacy, right?

[00:36:22] To, and there's a lot of girls there.

[00:36:24] I have to say, I'm super excited about that.

[00:36:25] I, you know, we had a group and I made two girls co-CEOs, 13 years old.

[00:36:31] It was amazing to see.

[00:36:32] And they took, they took that responsibility very, very well.

[00:36:36] And they committed national level, pitched in front of a lot of people, not being afraid.

[00:36:42] At this stage, I'm not sure I would have been able to do it when I was 13.

[00:36:45] So I was really excited to see girls doing this.

[00:36:48] We're going to have to talk more about that offline because this is exactly what I proposed to my daughter.

[00:36:54] My daughter's a junior in high school.

[00:36:56] This is exactly what I just proposed to her superintendent and board of education and superintendent of curriculum.

[00:37:03] Well, we're replicating this everywhere.

[00:37:05] So it's a great program.

[00:37:07] I'll send you the details.

[00:37:08] All right.

[00:37:08] Yeah.

[00:37:09] Let's talk about that.

[00:37:10] But yeah, I think within organizations, I think it's inevitable that there will be some sort of, just like we have training on when you onboard and then on some regular basis around, you know, data privacy and harassment and cybersecurity and all these things to try to mitigate, you know, weak links, essentially, and bad behavior.

[00:37:33] I think it's inevitable that we'll have some like responsible AI kind of module in there.

[00:37:39] And then depending on roles, you know, you'll have other, you know, training and upskilling, obviously.

[00:37:43] But I just think some foundational, you know, course, even if it's only an hour or whatever, it's better than, it's better than nothing.

[00:37:51] It's just so at least people will, will think a little bit more about the output of some of the things that they could be not just using, but building, right?

[00:37:59] They can build their own agents and co-pilots and GPTs now.

[00:38:02] So they're probably going to do it because even when we see adoption rates at companies like you were talking about before in the, you know, 25, 30%, maybe 35% range, this sort of shadow AI use is, you know, more than double that, right?

[00:38:18] If you ask the employees, it's probably like 80% have tried it.

[00:38:23] Whether they've tried it for a work use case is maybe different, but they're already playing with it, which is exactly what I told the school administrators.

[00:38:31] Like, you know, your kids are using it.

[00:38:33] If you tell them not to, then they're going to use it more.

[00:38:35] No, you can't.

[00:38:36] You have to embrace it.

[00:38:37] And exploration, continuous exploration.

[00:38:39] I have to say, I'm not keeping up with all the different things that are coming out.

[00:38:44] I'm super excited every time my son is coming up with, did you check this?

[00:38:48] I'm like, no, let's check it.

[00:38:50] It's a lot.

[00:38:51] It's a lot to keep up with, even with my probably 15 different AI newsletters.

[00:38:56] It's just a lot.

[00:38:58] Isabel, this has been really great.

[00:39:00] I want to be respectful of your time.

[00:39:01] I really appreciate you taking some time to talk to me and share your insights with my audience.

[00:39:07] So thank you again.

[00:39:08] Thank you for having me.

[00:39:09] It's a great pleasure.

[00:39:10] And definitely we need to connect on all the different initiatives that are out there in the tri-state area.

[00:39:17] Perfect.

[00:39:18] All right.

[00:39:18] Thank you so much.

[00:39:19] Sounds good.

[00:39:20] Thanks, everyone, for listening.

[00:39:22] That concludes another episode of Elevate Your AIQ.

[00:39:24] We'll see you next time.