Speaker 1 (00:07):
[inaudible]
Speaker 2 (00:08):
Hello and welcome back to another unleash webcast. This is George. Laraque the SVP of insights that unleash. And today we’ve got an exciting conversation coming your way, brought to us by eight-fold. It is unlocking the workforce potential with talent management powered by artificial intelligence. And to lead us through this discussion, we have Steve fare, the director of product marketing at eightfold and grant Weinberg, the VP of global talent acquisition, HR operations, and H R I S at Tri-Net. We’ll let them introduce themselves in a little more detail in a moment, but I’d like to cover some housekeeping with you before we get started. Uh, during this webcast, uh, you will have the opportunity to interact with each other in the chat window. Uh, we encourage you to share your LinkedIn profiles. Tell us where you’re joining us from. Uh, tell us what topics are of interest.
Speaker 2 (01:08):
We will be engaging with you through the conversation at the same time. You’ll notice next to chat on the top of your window. There is a Q and a tab. Any questions that you enter there may be selected at the end of the discussion for a question and answer period that both Steve and grant will be, uh, participating in. So get your questions in at any time. There’s no good time, uh, like the present, when you hear something that interests you also, we will have a handout for you and you’ll notice a handout tab, um, not surprisingly right next to that Q and a tab. And when you see the red, uh, light eliminate, there there’ll be a report there for you brought to us by eight-fold, uh, the ultimate guide to talent management in the age of artificial intelligence. Uh, it’s something that I strongly suggest you dig into and you’ll hear some of the topics covered in the discussion today, uh, with that I, uh, wish you both Steve and grant a fantastic, uh, conversation. Thank you eightfold for bringing us bringing this to us and our audience, and I’ll hand it off to you, Steve.
Speaker 3 (02:22):
Well, George, thank you very much and thank you everyone for joining us today. It’s a wonderful to have you all here virtually. So hear more about the unlocking workforce potential. Uh, certainly look forward to speaking with you about your questions at any time. So please do send us your questions and we’ll leave plenty of time at the end to take them today. I want to briefly talk a little bit about who we are at eightfold AI, uh, before turning it over to grant to introduce himself. And Tri-Net here at the eightfold. Heath invented the talent intelligence platform, which is the most sophisticated artificial intelligence platform for talent it’s been developed today. Uh, it’s so sophisticated. In fact, we’ve earned a patent in the United States for our algorithms and it’s been built, uh, by the true global experts in deploying this type of technology for organizations we’re in mountain view, California.
Speaker 3 (03:23):
Uh, we’re still con Valley based, although we are individually all over the world, wherever you are. We’ve been at this now for about five years and we’ve already signed up more than a hundred large enterprises and even some governments now, uh, on, uh, continents across the world, we’ve been impacting careers, uh, more than a million and a half careers to date in 110 countries, uh, and in 13 languages. So that’s just a little bit about who we are, and I want to turn over to grant now, uh, to, uh, introduce us to try that this himself,
Speaker 4 (03:56):
Thanks, Steve, and, uh, appreciate being part of today’s work costs. So my name is grant Weinberg. I lead talent acquisition, HR operations, and HR tech at Tri-Net. And for those of you that don’t know, Tri-Net TriNet is a professional employer organization. We provide smaller, medium sized businesses with full service HR solutions, and then by industry, uh, to free the SMBs from HR, complexities it off as access to human capital expertise, benefits, risk mitigation, compliance, payroll, and real-time technology. So from main street to wall street, trying to imply, empowers these SMBs to focus on what matters most and that’s growing their business. Um, we have been looking at eightfold for a while, um, and I’m excited to now be a, an eight-fold client. And so, um, back to you, Steve, as we, we get stuck into the session, but
Speaker 3 (05:00):
Well, thanks so much grant. So we love Tri-Net wonderful technology, really easy to use. So I want to turn to today’s topic, which is unlocking the potential of your workforce. And in order to really set this topic up, I want to talk a little bit about how business is changing and why the workforce is at the center of that. And something that we’ve seen at eightfold is how in the last few years, talent management has become very central to how organizations think about how to approach their individual futures. And by talent management, I’m referring to the function of keeping and growing the employees that you have within your organization. And this is not something that is an afterthought for companies anymore. It’s often in many cases become a C-level issue, even a board level issue because executives recognize that their people, their is why leading organizations become leaders and it’s widely to organization stay leaders in any industry.
Speaker 3 (06:05):
It’s all about the people. So how do you become a leader, or if you’re already a leader in your industry, how do you maintain that position in order to do that? You need to do what’s right for your talent. You need to think about talent management so that your employees stay and grow. If you take care of your employees, they’ll take care of the customers. They’ll take care of the business. And this is often described as the future of work. This process of making it work for your employees is very central to the future of work. And so that’s the biggest change that we see in business, but breaking down that change even further, employees are really demanding a lot of new things. They’re not asking for the future of work. That’s not the words that they use, but what they are asking for adds up to it.
Speaker 3 (06:52):
And in order to show why I want to share some stats that we found at eight-fold when we surveyed a representative sample of several hundred employees, and what we found really captures these demands employees, first of all, say that they’re not happy in their current role. A lot of people say they want to change roles, but they don’t all say they want to change companies. What they’re really asking for is a new challenge, new opportunity. They would rather stay with you, but they would like to learn a new skill or even have a temporary change so that they can come back to their full-time role, refreshed and motivated with their, uh, with their new skills, the new colleagues that they’ve met. Now, we also are under trying to understand what a workers based on their age are looking for. And we found that younger workers say that they’re looking for a purpose.
Speaker 3 (07:42):
And by younger, I generally mean the millennial and gen Y and gen Z workforce broadly under the age of 40 today. And they’re looking for both individual purpose, but also corporate purpose. They’d like to see that future in a different way, but in a very meaningful way, the older workers, uh, those individuals who are in gen X and baby boomers, they, they also say that they would like to continue to see that future path for themselves. But it’s, for them, it’s less about finding the corporate purpose and more about finding that individual desire to stay at the very forefront of their careers and what it overall adds up to is two and three workers looking for that clarity in their career. They wanna see a clear career path that matters to them and breaking it down further. We see that that future of work comes down in four separate buckets and especially important in this disruptive year of 20, 20 digital and remote ready tooling is critical because your employees may return to the office, perhaps they have already, but you’re still going to have employees who are in hybrid or fully remote roles, and they need to have the same ability to access career paths and purpose as anyone who works in an office or works on site at the same time, everything that you’re doing for your workforce needs to support your imperatives for diversity and inclusion.
Speaker 3 (09:10):
Also no longer an afterthought. What’s more everything that you’re doing to enable what your employees are demanding needs to focus on their experience. Uh, employee experience is a common buzzword. Yes, but it’s very meaningful. It needs, you need to connect with your employees where they are both physically and in terms of how they’re used to using tools and technology. And while you’re doing this, you need to recognize a reality about the future that we’re in today, which is that skills are changing very quickly. We find that the half-life of a skill has decreased to four years and continues dropping, meaning that the typical skill that you have four years from now, it’s as, just as likely as not that that skill will be obsolete, it’d be replaced by a new skill or a change to the existing skill. So this future of work is boiling down to making careers possible for your valued employees and to giving them a way to develop their careers.
Speaker 3 (10:02):
And in order to do this, you should ask yourself a few questions to address the needs of your employees in light of changing business goals that you have. How can you make sure that the demand that you have for talent and the capabilities of your talent is being addressed? Do you have the people or the people you have, have the ability to meet those goals, thinking even more accurately or precisely about those capabilities? How do you obtain them? And this is really important because it’s not just about hiring. It’s also about managing your talent. It’s very important that you’re empowering the employees you have. And that really does come down to the word empowerment. I use that word very consciously because it’s about giving employees that sense of control, that they are developing their own careers, that they see options, that they see their own future.
Speaker 3 (10:58):
Creating that future for your business is about creating futures to your employees. And so that’s what we try to do. And that’s the mindset we try to encourage for talent management, starting with the needs of your employees. You will achieve the metrics, the goals that you may be responsible for, such as retention, employee engagement, skills, growth, even overall productivity comes down to transforming talent management for your employees. That transformation requires giving your employees tools that do three things for them. First of all, these are tools that need to be self-service. Even if your employees are not going to be working at home, your employees are used to having always available technology and they’re used to access accessing at any time. They want to be able to explore career options. Anytime not depending on a conversation with their manager, not depending on a review with HR or a quarterly process, it’s about self service.
Speaker 3 (11:59):
Let me put this another way wherever you are right now, where is your smartphone? I bet it’s in your hand. And if it’s not in your hand, it’s in your field of vision that reflects the level of self service that you expect from your consumer experiences that you have on your smartphone. You want to think about having that level of accessibility, that level of self-service for talent management solutions. All right. So what else do you want to think about? You want to think about making the solutions based on capabilities. This is to address the fact that skills change so quickly, but I use the word capabilities to add another nuance. A skill, a skill is something that you can do today. You have a skill to do something. A capability is a skill that you could do today, or that you could learn very easily to do tomorrow.
Speaker 3 (12:53):
So a capability is based on learnability and this is also something that you can enable for your workforce, that they can see, not only what can I do today and what can we do with that, but also what could I learn to do tomorrow? And how can I get there? And this is all based on each individual. This last point is very important. It has to be personalized the employee. You’re not treating employees as numbers. You’re treating them as individuals who may have their own specific goals. They may not always be about your company specifically, but that’s okay, because when you personalize what you’re providing to each employee, you will still encourage them to pursue their own future based on those needs. So the type of digital technology that you’re looking for, the type of transformation that you need to have, the way that you unlock that potential of your workforce is to think in these terms. Now, I want to take this one step deeper and think about some of the specific challenges that you may be facing as an organization, as a talent leader and how you can apply this mindset, self service, capabilities, and personalization in order to address a very specific needs. And I want to start with one, which is okay, let’s be honest, it’s an uncomfortable conversation, but it’s one that happens all the time and that is workforce reduction.
Speaker 3 (14:18):
So it’s a reality. It’s a fact as business, that layoffs can occur, that whatever words that you use to describe them, uh, redundancy or workforce reduction, uh, reorganization it’s, it’s just the fact that in almost every business, at some point, there’s going to be an elimination of a position. For whatever reason it happens. It might affect just a few people at a time, but it also may affect divisions, departments and thousands of people. And the challenge that organizations today is that when reductions in positions happen, usually when that position is eliminated, the employee who holds that position also loses their employment. And that’s because organizations are not set up to address, move it from that employee, for that employee to another role, they’re not set up to address redeployment. And that’s because organizations just aren’t set up that way. The jobs are siloed and departments, job descriptions and skills are managed manually.
Speaker 3 (15:23):
And there’s a short timeframe for redeployment. And when it’s all manual, the HR staff that facilitates redeployment are in very short supply and are very limited, but it doesn’t have to be this way. This isn’t ideal because it’s expensive to lay people off very frankly, and it’s doubly expensive. If you’re just going to hire someone with the same skills to do this, to do different work. When someone in a role could have taken that job, what’s more reductions in force hurt productivity. They hurt morale. They’re very disruptive. If you’ve ever been in a workforce, that’s encountering a reduction in force, you know, this from your own experience. So one of the things to think about transforming your workforce and unlocking its potential is what if you could do it a different way? What if you could separate the elimination of a position from the reduction in force?
Speaker 3 (16:15):
And this is an approach that you can take when you have digital tools. When you have deep learning AI, you can match people to roles, scale based on their capabilities. And this becomes something you can actually do. So this enables you to take a slate of impacted employees and see any available, invalid job, anywhere else in your organization, that that person could take, that you can offer them options and do it quickly and redeploy rather than reduce to the greatest extent possible, which not only provides bottom line benefits to you, but also provides tremendous cultural benefits to the business. This is being done today. Uh, we do have a client that had to eliminate unfortunately, a several thousand positions and they were able to redeploy a far larger number of the impacted workforce that they were expecting based on their benchmark for past workforce reductions.
Speaker 3 (17:14):
The redeployment rate went from a benchmark of 19% to 58%, which represented many hundreds of people. So this is something that becomes a transformative process that is enabled when you are working on an individual basis and a capabilities based basis. And I, this is something grant I wanted to bring you in here because when we were speaking before you would talk about this flexibility of talent redeployment, and this is something that had really resonated with you. I wonder if I could turn it over to your grant to talk to us more about, uh, your thoughts in this area.
Speaker 4 (17:51):
Sure. Thanks, Steve. I think this is the power of a tool like eight fold. Um, you know, we all know that the, when the pandemic kids at the beginning of the year, companies had to pivot really quickly, um, and particularly companies that are, they have frontline workers or essential workers. And so part of our business does have a, uh, an essential workforce. And so it was one thing identifying who the essential workers were, but because of COVID and you know, what the pandemic was showing, we need to, we needed to ensure that we have continuity in that essential workforce. And so without eightfold, that process was very manual of, you know, pulling resumes, looking at people’s LinkedIn profiles and trying to understand who in the organization had the relevant skillset. Then, you know, if for any reason somebody did get sick, or if we had to quarantine a certain number of workers, um, the work still needed to continue.
Speaker 4 (18:54):
And so we had to make sure we, we were able to supplement the workforce. And so when you do it manually, as, as you cited all those challenges, um, you know, you require a workforce to do it, a team that can do that analysis. It takes time, the data is not always real time. Um, and so the accuracy of that, the work and the outcome is questionable with a tool like eight-fold on empathy with a button you’re really able to very quickly understand in the, in the workforce, those skills are past or present, right? We, we might’ve hired somebody into doing job on that pain from a very similar job that we need to, to ensure that it remains essential in our workforce. And so this is the advantage you get, where, um, very quickly and at scale, um, you able to identify the talent, um, and it just gives you a competitive edge, right?
Speaker 4 (19:51):
Whether it’s, it’s ensuring that you can keep the, um, the work, the work moving, or that you don’t have to go out and hire more resources, whether they’re contract or full-time exhausted resources and then increase your costs. So I, again, you know, we saw it firsthand at the beginning of the year. Um, and, and certainly, um, as, as I look at other opportunities in our business, we already, um, we only about a month into the deployment of a fold, as we thinking and talking about different roles. So opportunities, um, whether it be succession planning or anything along those lines, we can start to use eight fold, um, to help us understand who in the organization has the skills for a role before we consider that other options.
Speaker 3 (20:41):
Well, I really appreciate that grant. And I love what you just said about considering employees for other options and in particular mentioning succession planning, because there are certainly other situations where you need to understand what each employee is capable of and what they can do next. And that really takes us to, uh, something else that, uh, I honestly found very shocking when I read this report about the critical, critical importance of skills today in light of the fact of the half-life of skills is so short notice I’ve been talking, you’ve had a chance to read this short quote on the slide that Gartner found fewer than one in six new hires today, have the skills they need for their current and future roles. Meaning that very few of the people you hire, who presumably you’ve hired because they’re the best candidate for the job are set up to be successful in the longterm.
Speaker 3 (21:39):
And that just reflects how difficult it is to manage skills today and understand what each person can do. And, uh, be able to actually apply that information. I often hear from organizations that have acquired skills ontologies as a source of data, and they find that the actual delivery of this information just isn’t actionable. They’re not able to take this understanding of each person’s skills and actually use it at large scale. So moving beyond this takes that ability to understand what each person can do and then also their capabilities. So up-skilling, and re-skilling also becoming really critical a function that is enabled by deploying AI and by managing talent management with AI, it involves showing each employee the benefit of learning new skills, making it personalized to their goals, and then letting them do it on their own time and removing that manual effort, not just to the employee, but also from the HR professionals who are tasked with managing these programs.
Speaker 3 (22:49):
When you can do that, you’re going to have up-skilling and re-skilling programs that work, first of all, because employees will actually use them a really critical factor here, really critical factor is having that skills information built right into the platform. So you are not also trying to manage it individually or manually. The schematic here shows the way to think about this, the way that this can work are these squares and circles. You have an employee who is in a current role, and they may be looking at some future goal. It could be a promotion as is illustrated here, but it could also be a lateral transfer. They may just be interested in doing something different. It could even be moving to a different office or a different division, whatever their goal is, and they can explore many different goals. They need to understand how to get there.
Speaker 3 (23:40):
And that’s up-skilling, and re-skilling that works because they can see exactly which skills they have today. They can see what additional skills they may need to reach a goal and then understand exactly what they can do to gain those skills, whether that is a traditional course or a digital course student LMS system, whether that is a short-term project or gig in a project marketplace, whether that is a mentorship relationship. And w even whether this is a event put on by the company or an event put on by a third party, for which the information has been, uh, plugged into the technology. And these are all examples of, of, uh, ways to learn that April provides in April talent management. And of course the employee can not only see options that what they need to do, but also the options they have to get there. So this was a way to get to that usage and contribute to that upskilling goal.
Speaker 3 (24:40):
This is bottom up for the employee based on their own goal. This can also be driven by your organization from the top down, telling each employee here is a path that is defined for you, if that’s what’s right for your organization, and then simply making it something that’s much more self-directed on their own time, you’ll still see much more uptake and, um, uh, much more excitement about the process. So a grant, I wanted to turn it back over to you again. Uh, this was something that I know you’re very passionate about the upskilling and reskilling. Um, how are you looking at this topics around skills development and, uh, grooming all of your talented colleagues for their next role?
Speaker 4 (25:24):
Yeah, thanks Steve. Well, fortunately we we’ve got fold now, um, because it’s making life a lot easier for us. Uh, you know, as we all understand when it comes to skills development, um, you know, really you have your, your job architecture, you have job leveling, you have job descriptions, you have individual development plans and that, you know, historically that data has been all over the place. And so it’s been hard to consolidate it and really make it easy for, um, your, your employees to, to either develop their skills or to re-skill themselves. And so what we looking at with, with eightfold is really the ability now to say, you know, for certain, um, job levels or promotion criteria, here’s what the expectation is that somebody is proficient in these areas. Um, and that now allows the employee to earn their career development, um, because they can go in and they can upskill or re-skill themselves based on the information that is much more transparently, um, represented by the organization.
Speaker 4 (26:30):
And so I think what we will start to see is, um, employees owning their careers, um, and being much more the forefront of understanding what it takes on both from internal perspective, but also calibrating with external, because, um, for those that, those on, on the webinar today, um, eight-fold offers you that benefit of, you know, not only calibrating external candidates against, um, the criteria for the role, but if you’re using the talent mobility portal, you can put your internal talent in as well. And so, but you know, at a very, at the click of a button, we can see where the opportunity is for our talent. So can they make an own that responsibility, which I think just creates a much more engaged and optimized workforce. So back to you, Steve,
Speaker 3 (27:24):
Well, thank you grant. I want to emphasize, as we were talking about this, this really critical importance of doing it for everybody on the workforce and making it accessible to everybody, I think this is another important point to highlight, which is making that personalization front and center. And that aspect around, I mentioned diversity and inclusion at the beginning, uh, really making it part of what each employee can do. You can certainly have your own policies in place. Uh, you have the job transfer policies, you have requirements that any employee who’s met, those requirements should have the same opportunity as any other employee to look for internal mobility options, uh, to, to do this, to create this expectation that each employee will want to look internally first, before they look outside. Because I think that’s the reality today is that when an employee is says that they are looking for the next thing, all they’re going to do is begin looking for another full-time job and quit.
Speaker 3 (28:33):
We want to interrupt that cycle. We want to increase retention of productivity by giving employees other ways to scratch that it should do something new and making that available to everybody. Um, one really quick story about this. Uh, one of our, uh, other, uh, clients had been, uh, looking to make this possible. They did roll out, uh eight-fold to their organization. Uh, about 6,000 employees in total were receiving this technology. And after 12 months they did to come back to measure internal mobility. What they’d found was that 16% more employees were moving between departments and 25% more employees overall we’re switching roles. And this was just full-time opportunity. There’s also opportunities for many other kinds of actions. Employees can take such as mentorship and projects and even learning opportunities that are connected to a specific goal. It just becomes an opportunity for each individual to find their own path for mobility. Um, grant, I know you had had some thoughts this as well, when we last spoken. Uh, did you want to add any other comments around, uh, how, uh, thinking about mobility for everybody, uh, impacts your workforce?
Speaker 4 (29:53):
Yeah, absolutely. Absolutely. Steve there for us, the, the mobility opportunities is historically it’s been held by the line manager. The line manager will observe the talent in their organization, and they will decide who they think is ready for either promotion or opportunity elsewhere in their organization or potentially outside of their organization, but still within the broader organization. And so we, you know, one of the, the advantages we saw with eight-fold is, um, w from a diversity and inclusion perspective, that process that I described, um, can have some biases in it, right. Uh, you know, I aligned with people that I know, I like, you know, I perform in a certain way, some of my team performing that same way, they get preference over others. And so by deploying eight fold, what we want to do is we want to do put that career mobility in the hands of, um, the employee, um, and, and help support the hiring manager.
Speaker 4 (30:59):
And so what we now have is, is it two way facilitation out? You know, we encourage our employees to go in and fill out, um, their profiles on, uh, eight fold. The machine does its work, and it starts to create career paths for them and shows them what the possible paws internally are. And that’s a really important point, right? We, you, you mentioned it earlier, Steve, um, we don’t want people to go and start a career search on an external platform. We want them to be able to identify what they, not only their next job, but probably the next two or three jobs, or even make a career shift internally at Tri-Net, um, and, and go into a different functional area. And so by using the machine learning and the technology to help us identify those pods, um, we now have, have helped the employee understand at their direct access what the, the career path looks like.
Speaker 4 (31:59):
In addition to that, um, the system helps identify identifies, um, mentors, which we all know from a diversity and inclusion perspective is very important, um, for the underrepresented groups. So now not only do we have possible career paths, we also have people that can help, um, the, these internal employees understand what it takes or what they need to do, um, to, to make that next move. And then to underpin that, or, um, eightfold is connected into our LMS is. And so those learning management systems will give a high level view of potential, um, and academic opportunities that the employee can take as they close the gap on their skills and experience towards that next level. And so we now have a situation where the employee owns that process of understanding their career paths and the next steps they can meet with a mentor and understand how to get there, or how that individual is not there.
Speaker 4 (33:02):
Um, they can take additional training and upskill themselves. And so when it does come to promotion time, or just your end of year review, or even, um, an employee looking at another internal opportunity, there’s data to underpin that move, um, and it also supports, uh, by the HR and the hiring manager in making certain decisions around certain talent. And so we’re very excited that, that the technology is going to help us, um, not only remove biases, um, but really unleash the power of our employees, um, in owning their careers and, and, and growing with us as opposed to break outside of the organization.
Speaker 3 (33:48):
Well, thank you for that grant, the unleashing, uh, employees, that’s, uh, a wonderful concept. Uh, I think, uh, wonder if others have thought of that before I, uh, I want to, um, give us a chance to talk about, um, an important aspect of this, as we’ve been talking about where organizations should go with talent management and how artificial intelligence is a critical, critical component of this and how artificial intelligence is a critical underlying technology. Um, so I do want to, uh, take just a few minutes to mention some key considerations when you’re looking at artificial intelligence. I know if you’re, you’re listening to us, you’re thinking about artificial intelligence, hope you’ll consider eight fold. Certainly whichever AI you consider, there are some important points to understand what you’re getting and what that AI can actually do to see that if it’s going to have that ability, uh, to unleash your workforce and, uh, truly get you where you need to be.
Speaker 3 (34:55):
Um, as I switched to this topic, I do want to make sure if you have any questions, uh, we’ll take your questions live at the end of the webinar. So we’d love to get your Q and a, um, so that we can not answer what the additional questions that you have as we go through this today. So, uh, please, uh, send us your questions. So there’s a lot of things you’re going to ask. Of course, you’re going to have a questionnaire for the vendors, but I think there’s three really important things to ask about AI, to make sure that you understand what you’re getting at first is you want to understand what kind of AI, what your provider means when they say AI. Secondly, you want to make sure you understand what that AI is going to do. A grant has talked a lot about the data, and that’s so critical.
Speaker 3 (35:42):
What’s the AI going to tell you that you don’t already know AI does not replace humans. AI doesn’t make decisions. AI makes predictions, make predictions with maps. So how can I be confident in the AI predictions to make sure that you and everyone using it is going to understand what is coming from that AI. So first, what does the provider mean when they say they have AI? We do hear a lot about AI. It’s a very commonly used term today, but they’ll often that usage. Does it answer this question of just, what is it, what is AI? And as I said, a moment ago, AI is math. Uh, AI is a set of statistical calculations. Typically it’s not a robot like you might see in a movie, and it’s definitely not, uh, intelligence, like in the way that you would say a person is intelligent.
Speaker 3 (36:36):
In fact, if the world’s best AI isn’t as intelligent as an animal because it’s math. But what AI does simply is to give you that prediction, to give you that response based on an input. And that’s very useful, it seems insightful. It’s valuable and can even seem like a real intelligence. All right. So within this world of AI map, as you see on the screen here, there is a hierarchy of methods and others may define them a little bit differently. I’ve, I’ve seen several different models. This one’s pretty widely accepted, uh, hierarchy with artificial intelligence is the broadest terminology. Um, machine learning is a subset of AI and deep learning is a subset of machine learning. And what I would encourage you to do when you’re talking to any provider of AI for any use, frankly, understand what you’re getting.
Speaker 3 (37:25):
Um, this last type of AI deep learning is what eightfold uses. Um, it is the most advanced kind of AI and for talent management uses, uh, certainly believe that it is the type of AI that you would need. There may be other uses for which a simpler type of AI is appropriate. So, um, I can’t, uh, in the context of today’s call, we don’t have enough time on this webinar to give lots and lots of detail, but just to give a very brief description of what the highest level of AI does, it gives a response based on an input, but Bunce, once those responses are programmed, they don’t necessarily change. They don’t change over time. I should say. So a decision tree, for example, is a type of AI at this basic level and, um, decision trees, given output based on an input, but that doesn’t, doesn’t change over time.
Speaker 3 (38:19):
This type of AI requires maybe high school math. And it’s actually, I think how a lot of us learned to program, um, if then else statements, for example, uh, personally, I never got beyond that in programming, but that’s, uh, that’s this level AI now machine learning, it’s a subset of AI, which is more advanced. It’s a system that’s going to use the statistical fit method to give the best response based on a certain input. So in order to give those right responses, machine learning system needs training data. It’s going to need typically thousands of observations. So thousands of events, where we had these inputs and it gave this output, give this prediction, the machine learning system can then use ongoing human based guidance to fine tune its model and become more accurate over time. So this is harder than just the basic AI techniques, but the reward you get is a system that could be improved over time by human operator.
Speaker 3 (39:19):
Let’s say this is usually a college level or even master’s level math and statistics techniques, and deep learning is the most advanced pipe. This is the most difficult to create, but also as the most reward at the end of it. And this is going to use statistical methods based on a type of computation called neural networks, which uses multiple layers of computation. Uh, deep learning systems need a lot more data than machine learning systems. They need millions of observations of training, data and feedback. The data automatically become more accurate over time without further human intervention. This is very difficult to program and create this create. This requires masters and PhD level folks. So typically in order to do this, but as I say, the benefit you get is a great accuracy and flexibility. So lots of organizations can create AI. In some cases, a simple decision tree or other simple technique may serve your needs.
Speaker 3 (40:19):
But if you have complex large organizations with large needs, make sure you’re getting AI that actually can meet that value and meet that need something else that you want to think about is that the AI can, uh, give you, uh, understand what it’s providing you, what additional information it’s providing you. And one of the things that should be able to do is help you further understand each individual. This is something that we are certainly doing with eight-fold. Um, but again, there are other solutions that can do parts of this. You just want to make sure you understand the detail, what you’re getting what’s highlighted on this visual is examples of the additional information, uh, that April’s able to provide on a, uh, on a profile. So for an individual, for an employee, we understand how much experience they have is relevant. We understand the context of the companies and jobs that they’ve had.
Speaker 3 (41:16):
We’ve understand whether their career is, uh, is, uh, has notable experience. In other words, are they precocious in their career advancement? We can understand whether they have similar profiles in terms of the career advancement to others, and we can understand the skills they have and the skills that they can learn most easily. And I want to highlight this as a really important area of this enrichment. AI should give you this ability to understand capabilities when you’re using AI to unlock the potential of your workforce, because potential is really about what someone can do, not just what they can do today. And this is core to our approach to AI. We don’t just have a static skills ontology that’s that’s not going to give you that transformative need. We have a capability matrix that updates automatically and focuses on. Learnability not only the skills that the person has today, which we can detect automatically, but also what skills could they learn, which we can also present and also show people how to learn them.
Speaker 3 (42:21):
It does evolve over time as well. So as the world continues to change with this very quick half-life of skills, we can continue to change. As we see, um, the career paths of millions and millions of people across the world and understand how people’s skills, not just in your organization, but in every organization are changing over time. Uh, in total, our capability matrix knows 1.4 million unique skills. There are certainly other, other possibilities out there. So one thing I would encourage you to do is understand how the AI you’re considering helps you understand greater context than just what is in the profile you have in a particular understand the potential of each person.
Speaker 3 (43:07):
Now, one more thing that’s really important when you’re considering AI is, uh, the confidence you want to have every single person using it, understand that they can trust the predictions that they get. And really AI can give predictions that are very strong, much stronger than a human based on the data that they have, but in order to get people confident that they need to see why. And that is a very important part of what AI can do for you. You want to make sure that there is explainability. This is something that we do as well with our AI. Every prediction has a visual that shows why that prediction was made, whether it’s based on skills, whether it’s based on companies and titles, how that information combined in order to give the prediction. So this is something that you should expect from any AI that the information about that prediction is made available to the decision maker.
Speaker 3 (44:07):
Uh, finally, something else that is very important is understanding how that AI addresses issues of bias. Again, this is a topic that is a very important, I can’t do justice to it today in terms of, uh, the, the depth of this topic, you want to make sure that the AI, anytime of Wrightsville recommendation also is provided that recommendation based on the factors that matter such as a person’s skills and their experience, and that it is not based on factors that don’t matter. And that’s personal characteristics. The skills that someone has mattered someone’s age does not matter the degree that that person got at school that may matter someone’s sexual orientation does not matter. She want to be very confident that you have AI that is unbiased. Uh, at eightfold, we use a methodology called equal opportunity algorithms that tests our algorithms continuously to make sure that the predictions we provide are unbiased at a statistically significant level based on both known and unknown factors.
Speaker 3 (45:18):
So only the factors that you intend to such as skills are being used for the prediction. We do apply this rigorously to make sure that we’re not providing bias. Um, personally, I am very pleased to see that other AI providers, not just in the world of talent, but across the world are also taking bias seriously and are using other methodologies or this methodology, but whatever they’re doing, they should have an answer. And a way to show that they’re preventing bias inspecting this method may, may require somebody with a statistics and data science background on your side, but bring them in because they’ll want to see this. So, uh, I thought grant, uh, you wanted to, uh, add a couple of thoughts about this process of gaining buy-in for AI at Tri-Net. Can you tell us a little bit more about the conversations you had about AI and how you were able to, uh, win over your audience and before turning it over to you? Just want to give everyone one, uh, one more quick plug, please give us your questions. Cause those are coming up next.
Speaker 4 (46:27):
Great question. Um, you know, one of the, uh, as I look to fold, and this is going back to 2018, so you guys were coming up to your second birthday. Um, it was new technology. It wasn’t known technology, um, in, in the TA space. Uh, and so, you know, as AI became a, uh, a bigger buzz word, the opportunity to bring in such a forward thinking technology is, is always a nerving, right. For many in the organization. Um, and so as I sat down on this journey, one of the things I I’ve paused and thought is who do I need to bring on from, you know, from a collaboration point of view, so that as we go through the assessment of the tool, um, any questions, especially around this, this new, um, buzz word of, of AI would be answered and not become an, uh, an objection or just a lack of understanding.
Speaker 4 (47:27):
So it, it becomes something that somebody can support. So I did two things. Um, I, I looked at bringing in, um, you know, somebody from our legal team, somebody from our employment standards team and somebody from our technology team. And by having this representation, what I was able to do is to get, um, that kind of concerns about the tool onto the table and allow eightfold to really come to the forum and explain, um, how the algorithms are regularly tested and, and don’t, you know, develop a bias, which then gets replicated. And so from the, the assessment and due diligence of the tool, there was a lot of rigor around understanding, um, the, the, the, the AI, what I thought was interesting as we went through the process, um, one of the stakeholders shared with me, um, but they didn’t know a lot about AI, but they were just, you know, what they didn’t know about it was, was, was something to give them concern.
Speaker 4 (48:29):
And so before we had them, they mentioned to me that before they had the next review of the AI and getting deep into the algorithms actually took a couple of online courses to just understand the fundamentals of AI. Um, and, and, you know, I guess test some of the concerns that they ha they had answered as we went into that next meeting. Um, that individual was AP was able to have a much better open mind towards, um, the tool, uh, the use of the tool and, and, and, and, you know, allay any concerns that they had. We continued to keep that whole process, um, as, as part of, you know, implementation of the tool. And so, you know, we have through our, um, our implementation through our change management, uh, we’ve kept, um, basic links to, um, videos on YouTube that explains, um, AI either through eight-fold or through other sources, so that anybody using the tool, um, now that it’s been implemented as the opportunity to go and understand the AI.
Speaker 4 (49:42):
And so that’s been that kind of change management piece and understanding as we bring the tool into the organization. On the flip side, we have a lot of enthusiasm, right. Um, you know, I don’t want to name other tools, but, you know, we’ve used some legacy tools in the TA space. And so eightfold has been a revelation to the users, um, whether it’d be the talent acquisition team, the broader HR team, or even just our, our employees. And so what, um, we’ve been very clear with, with, with the users is to explain that AI is not in, you know, it has a level of intelligence, but from day one, it’s not going to give you the outcome that you’re looking for. There is still a need for that human intervention to help calibrate and help teach the, the, the AI to be, um, more effective and really understand your needs and validate your needs.
Speaker 4 (50:37):
And so, you know, each one of these interactions is, you know, somebody rings a member of my team and says, I completed my profile. I’m in finance, and I’m being served up jobs in, in as junior payroll jobs. The system doesn’t work. What we’re helping those people understand is, um, today it’s, it’s kind of saying that you’re, you’re, you know, you could look at a payroll, uh, job as we calibrate your experience in this, as we calibrate different roles within the organization. And that gets back to, um, you’ll start to notice that the machine will get smarter. And as we know it about the six month Mark, we, we anticipate those that those are anomalies, um, w w will not show up any longer. And so the importance of continuing to help people understand AI, to take away any, any kind of trepidation they have.
Speaker 4 (51:26):
And on the flip side, as they start to use, just reminding, uh, reminding them that from day one, you know, we don’t have a super computer, it takes time and it takes collaboration of the technology and the human to get to, to that point. And so I’m sure this is a journey and what we’ll continue to focus on it. Um, but for anybody out there that is considering a tool like eight-fold, I would highly recommend that education around AI and deep learning beyond just the buzzwords. Um, you know, a lot of people out there, um, have the understanding level at, so back to you, Steve, unless you have any questions on that piece.
Speaker 3 (52:06):
Well, I think our audience has some questions. Granted, if you can stick around for a few minutes, uh, love to turn it back over to George, to moderate some questions from the audience.