AI Works Best When Humans Set The Rules with Olga Rankin, Head of HR Arinco

Hi everyone, welcome back to another episode of the HOR Community Podcast with myself, Shane O'Neill. Today we have a very special guest, Olga Rankin, has joined us. Olga is the group manager for people and culture at Arinco. Good morning, Olga.

Good morning. How are you, Shane? I'm very good. Very good.

Thanks for joining us today. Just for our audience, Olga and I have known each other quite some time. So this episode has been uh a little bit of time in the making. As a background for those of you who don't know, if you're not involved in the Sydney HR scene, Olga has been within recruitment in HR for many years now, with a big chunk of that within the technology sector, has been part of the before dot com boom and is currently working through quite a an interesting time in tech with AI.

And speaking of which, Arenko are in AI-led technology business. So you're you're right in the center of it all, Olga. I am indeed, and I'm very pleased that you didn't say exactly how many years I've been around the uh recruitment and HR space. But I have been really fortunate in my career to go through and see businesses evolve and change through many different stages in technology.

And Arinco is the artificial intelligence company. We are an organization who are now in the in really the midst of AI and what it is affecting within the business world. It's really exciting, actually. Interesting enough, I guess on the AI side, and you and I spoken about this sort of offline as well.

There's a lot of noise in AI and technology land at the moment. And I think off the back of a lot of the HR and and talent events, it's it's it's everywhere, it's in every pitch, it's on every platform, it's in every conversation. When we're speaking to our HR leaders, there still seems to be a lot of uncertainty where potentially the risk sits, what's working, what's not. But I guess my question to you really is, you know, where do you think HR leaders are are maybe getting it wrong right now, or or maybe they're getting it right?

It has one really big thing in common across every industry, and that's people. And we're talking about using technology to remove some of that humanness from a process. And I think that there is lots of opportunity for us to be using AI, and I know there's lots of research about AI and AI interviewing, but I think the thing that we all need to think about is transparency. Actually, you know, most HR leaders are pretty busy, they're running pretty hot.

One of the things we need to think about is are we actually taking the time as leaders to stop and learn and really understand the capability of AI because we're all being told use AI, you need to use AI in your job. Are you using AI? Where are you using AI? We're all kind of using it, right?

But to be really using it effectively, we need to take a just take a beat and educate ourselves because we really run the risk if there's not guardrails in place for our teams, if we haven't taken the time to learn that, we run the risk of our teams just going a little bit, you know, off-pist and going through some nice and finding tools that they're like, this is a great tool, and it probably is, but it has, you know, maybe a lot of bias in it. Or it doesn't protect data, or our teams are using their own versions of a sanctions tool rather than the enterprise version.

So there needs to be a lot of conversation about it. And I know in my team, you know, we talk about it all the time about how we can use AI to enhance our roles rather than do our roles for us. Love that. That's great.

You touched on something there as well, which is around the learning side. I think, as you know, I'm quite a big advocate for AI and have immersed myself in it for many, many years now. I've noticed, especially in the last two years, there's been a lot more conversations around AI. And I can see there's different camps.

There's people that are people and companies that are quite progressive. I mean, the progressive ones now are they're so far ahead where you know a lot of us are actually learning from what they're creating and and and adopting. I guess where I would maybe see myself is still learning and applying and adopting. And then there's another camp which I think is still quite conservative, but I've noticed they've changed a little bit as well.

And I guess that it's like that whole change management persona where a lot of this camp initially thought it was maybe a fad, or or you know, it would eventually phase out. Whereas I think they're starting to realize, oh, actually it's not going anywhere, and we're actually falling behind significantly. So you can see those camp start to invest a little bit more in learning now too. I think that's 100% right.

I think there's a lot of people that are going, well, I don't want to do it because I'm a bit scared of it, a bit frightened that it's going to take away my job, or how good can it be? Or people won't want to deal with AI. And you're right, those people are starting to go, well, geez, it really isn't going anywhere. And actually, when I think about it, I'm probably using it nearly every single software tool that I'm availing to myself.

So I think they are on the way forward. And I think for myself, you know, I work with a a whole swath of technologists, and I spend a lot of time listening to conversations about AI and AI projects and what's happening and where they're putting in projects in different businesses. I spend a lot of time educating myself. I spend a lot of time practicing.

I practice my prompt engineering, I practice doing a lot of things because AI I'll give you a great example for HR. I could put in to an AI tool, here's my contract for my employee. Can I terminate this person? And AI will probably tell me yes, you can, because it will read my contract and say, yeah, there's a clause in there that says it can.

But it doesn't, because I haven't asked it, how do I do it? What steps do I need to take? Is there any rules that I need to be aware of? What are some pitfalls to watch out for?

I've just been told yes, because that's the question I ask. Can I do it? Yes, you can. But what we need to think about is asking AI and practicing how we want to ask questions, because our tools are only as good as the information we put into it.

We have to train our tools. And if we are not asking the right questions or asking it to think in the right way, it can't do it. I'll tell you what happened the other week. I thought I was so clever.

I was using Copilot, I was using Copilot Create and I have the premium version of Copilot. I'm lucky enough to have that license, so I'm in a safe space to use it. I put in our company logo. I said, using this logo, do blah, blah, blah, blah, blah, blah, blah.

Well, AI felt like our logo needed some improving. So it changed our logo. Not enough for me to notice, but enough for our marketing team to notice. So I have to learn.

I need to tell AI, don't change this image tool that I'm putting in. Now, you know, so it's a learning experience, but I'm also using a safe space to do it. I'm using a licensed version of an AI tool, of an AI platform. I think that is like the thing that I really worry about is somebody taking CV that they want to format and just putting it into a version of Chat GPT and the PII implications of that for an individual and for a company.

And those are the things that worry me. I love the questioning side of it. I think that's so critical. Another thing that I'd sort of learned with my prompting over time is giving it actual good context.

Like even if I take your example of look at this contract, can I fire this person with a little bit of context, you'll get a much more thorough, polished response that you can actually think critically, okay, is this going to be a good move or not? Whereas, like you say, if you if you just feed it in and say yes or no, it's gonna it's gonna come back with yes. We used ChatGBT for years, even in our business for a lot of things. And we moved over to Claude maybe six months ago.

And even I'm seeing the difference in how both operate. I feel like Claude is a little more polished, probably better suited for corporate. It probes you and quizzes you and questions you on certain tasks that you might ask it to support you with. We just done a big SEO project with Claude recently where it's done a full audit for us.

Whereas ChatGBT, I feel like it reminds you of someone or something that just kind of wings it sometimes. It'll just it'll just go in and tell you what it thinks you want to hear. Whether that's correct correct or not is another question. I often use different platforms to test which version of or I'd like to use and what's better and what's fit for purpose.

I think that's something that we also need to be aware of as HR leaders. The different platforms provide us with different ways of working. And we need to work out what's best for our organization. We also need to work out it depending on the type of organization you're in, platform might not be sanctioned by your organization, even if you think it is best.

So if Claude isn't allowed where you're working, you need to then go, okay, well, what can I use? And again, how can I get the tool and platform I'm using to give me the results that I need? So you could actually say to ChatGPT, I need you to act as an expert and using professional language and it will change the way that it presents information back to you. But you need to help it learn what you need.

But I still think that I use AI on a daily basis, but I use it to enhance work. And I think when we look at some of the things that are happening in the world, I can see where AI can provide huge impact to HR teams depending on the type of work that they're doing. So we know, for example, that I don't know what gen we're up to. Is it gen A and Gen Z?

Whatever gen we're up to, they prefer to use AI for some of their interviewing. And that might be okay depending on the role that they're doing. They prefer it because they can do it at a time that suits them, they prefer it because it's not intimidating, because they're not talking to a real person, so nobody's judging them. And unfortunately, in the world where I work, we're judged all the time.

And we're not judged for criticism, we're judged for growth. And and I think that's probably true in most projects. You're judged on your output. 100%.

And so it's okay to be judged to view a little bit, but there's lots of things AI can do to enhance our the grind, I which is what you know, those low-level noisy tasks that if you could just get those ones out of the way really easily, if we could have an AI tool that did that for us, why not? Yeah, I'd agree with that. And on the interviewing side, actually, I mean, we use tools. I'm sure you do too.

I'm sure a lot of organizations do. I know a lot of organizations that don't and that are probably listening to this with their pen and paper, but for you know, the screening, the assessments, even interviewing now. I know you touched on more of the candidate side there, but like fully AI-led interviews. I mean, I'm definitely pro for it, but but still sort of understand there are some risks to it too.

I mean, what do you think is the real risk, Olga, with fully led AI interview processes? And are HR leaders underestimating this a little bit? I think we are a little bit. And and again, I'll go back to saying that an AI tool is only as good as the information that we have provided it with to learn from.

So I think if we are ensuring that the tool has the context, has the rules that we need it to have, perhaps it can be useful at the beginning. I still think that and we also need to be very, very careful of the biases that our AI tool can develop over time because it will see our trends and it will think our trends are reality, whereas they probab they're not reality most of the time. Yeah. You know, so if, for example, a particular role you needed to have a CPA, but then the next time you're re recruiting for an accountant, you don't really care if they have a CPA, but your system, you haven't reminded it of that, so it's it's discounting everybody.

I understand who isn't qualified in that way. Well, that's gonna give you bias, it's gonna cut out a whole lot of people, it could I I think potentially lead to some questioning of your system and what you're missing. I still think people are an essential part of the loop, even if it is auditing some of those interviews every now and then to make sure that the system is working correctly. I and again, I think it matters hugely on the level of the role.

I agree. I probably wouldn't do an AI-led interview for a CEO. Yeah, I agree with you as well, because any of the tools we use for sourcing and a lot of that sort of heavy grunt work at the grind as you refer to it, we've set it up and built it in a way where it's essentially what we would typically look for. Like, so these are the kind of candidates we would typically look for for these roles.

It's essentially operating the same way we would, just at a much faster pace. So I think that that works for us. Where it also works really well. We worked with um an organization where they they had 10 different blue-collar trade roles that they were recruiting.

They typically get a lot of overseas candidates for these roles and they sponsor them and and whatnot, because they're on the skills basis, so or the skill skills um nomination, you know, another list I'm referring to. We found some of the AI tools from screening and sourcing and interviewing really helpful there because we could help them with recruitment operations and campaigns overseas, different time zones, but then get it to a stage where a human has to step in and assess before I mean before you start sponsoring and bringing people the other side of the world with their families.

So like it it helped in that regard. Otherwise, we would have been working through the night and to try and align time zones. Absolutely. And that's where AI, you know, when we talk about AI working in the background, that is a great example of it.

It's providing you with the labour force that you need to do something at a great pace and at great volume. Yeah. And in that case, I would imagine that there would be lots of rule in and rule out factors that AI could help you with because they would be, you know, the tickets that they have, the uh, you know, do they have police criminal charges against, okay, we're not going to be able to sponsor. So it'll be that would be quite useful, I think, a really useful use of AI.

And I think there are some other things that AI can be doing within that process and within that, within that interviewing process where it can go down different paths of questioning and almost getting harder and harder as we go along to to test the knowledge. But again, I use it more to do repeatable tasks that uh carry low risk. Yeah. For me, I feel more comfortable with that right now.

And I think that's the thing that we need to look at with our HR leaders. What tasks can we can we offload that are repeatable, safe tasks if we're not still 100% comfortable? So I'm very comfortable with AI doing lots and lots of things, but there's some things I still like and enjoy doing and take value as part of my role. And I don't want to give them up yet.

Yeah, 100%. 100%. And I know we talked a lot about where you and I have sort of seen improvements in our work from a productivity perspective with AI, what what's made some things better, improved some efficiencies. Have you experienced or seen anything where AI maybe isn't working or making things worse?

If there isn't enough context, I think that makes it worse. I think if people are not following the rules where we implement AI tools, we go to organizations and help them in their AI journey. So for us, having rules in place, ensuring the sovereignty of data is very, very important. Ensuring that there are no breaches is very, very important.

So I think the biggest downfall HR leaders are seeing is organizations not educating their people enough as to the why. Why must you use our version of a tool and not whatever you feel like using? Yeah, yeah. Now I don't know of anything.

I'm using the word of a tool here, could be wrong, but let's say Grok. Grok, great. Lots of people love it. Is it a safe environment for people to be doing work in?

Probably not. And so when we think about that, we need to be able to say, okay, this is the reason why you can't use that tool, because A, B, C, and D, and we've seen these things happen. I was listening to somebody talk the other day about a scenario where the AI agent was trying to blackmail the AI user. Oh, wow.

And I was like, oh, that's scary. That is very scary. Terrified, in fact. Wow, that's interesting.

Another area which is not a new sort of concept, but something that you and I talked about, I know you're quite passionate about, is that intersection between employee experience and customer experience. I mean, again, given the space that you're in, Alga, like what are you sort of seeing from an AI-led organization? Are you seeing AI play a role in that intersection with employee and customer experience? Definitely with employee experience, I think in our space, everyone's very excited about AI.

And there's a lot of education and a lot of training, and we have Teams channels that are devoted to what agents have been made, what great prompts have been created, and it's adding value to the work that our teams are doing. Our consultants at Aurinco have had to really lean into AI and work out how it can enhance their work because we are going to be seeing, in my opinion, real shift for HR, particularly in the L and D space and particularly in early careers employees, because those grind roles aren't going to be there anymore.

They don't need our our big four aren't going to need a whole fleet of early entry employees to do audits. Yeah. Because AI will take away some of those tasks. So how are those people going to learn how to work?

Because that's how we traditionally taught people how to work by coming in and doing lower level tasks. And if those tasks aren't there anymore, what are we going to do as HR leaders to give that customer experience and to give that employee the best experience? Because they're not going to have had the setbacks. They may not develop the resilience.

They may not have the core understanding of the why. And we need to have a really hard think about that. Because I think that is really one of the big challenges that's going to face our industry. Oh, massively, yeah, massively.

I mean, you would know at a lot of the events where we've sat down and had breakout sessions, you know, talent and skills always comes up at Top Trumps as a critical area for organizations. And, you know, one side of the room, you've got those that are industries where it's talent short, agent workforce, really, really struggling, you know, frontline healthcare, certain trades. But then you look at some of the sectors where we operate, where it's more service-based, consulting, recruitment, etc.

, AI is starting to really replace a lot of the roles, role duties, I should say. So we need to be ahead of the curve to actually go, okay, these are my areas of expertise, these are my skills. How can I leverage AI to, you know, I guess recreate or create a position for myself? Because the traditional way of recruitment, we're most likely never going to go back there again.

But I think as a traditional recruiter, your expertise are in your critical thinking, your ability to work through, you know, behaviors throughout the process, managing negotiations and offers, and you know, all that kind of stuff that you would know. Whereas AI can really do that as well. I mean, specifically in your sector, I'll get Is that something that's being discussed as well from a talent perspective, like how AI is affecting talent and what potentially that will look like in consulting?

Yes, we are talking about it because it's a problem, not a problem, it's a challenge we're we're working on solving. I have a large team of internal talent acquisition specialists, all of them working on ways to enhance their role so that they can work on the what we would call the higher value tasks. And the higher value tasks for us are all the interpersonal tasks. There's plenty of work that we are leaning into AI to do some some lower value tasks, but I think some of the roles are going to have to evolve pretty sharply.

And I'm seeing those those shifts in our behaviors happen really quickly. One of the challenges that we have, because we're a professional services company, so I could turn around and there's I'm surrounded by experts that can do tasks for me. However, they're all very busy doing stuff for our customers. So I spend a lot of time working with my team saying, well, go sit next to someone and learn it, because we need to know how to do this for ourselves.

That's one of the things that I think is going to come out of this is HR is going to play a bigger and bigger role in how AI affects our workplaces, what that's going to look like in role creation and role evolution. And we really need to be talking to our people, our our cohorts of employees on a regular basis about how they are taking charge of their careers and their learning. Because if they're not, I do think they're going to be left behind a little bit. Yeah, I'd agree.

And even with a a trade. Okay, so we could have you a trade could create a whole bunch of agents to do all of their administrative work. Now that's amazing. They could do all that, they don't have to pay a bookkeeper anymore.

So what's the bookkeeper gonna do? We need to work out how the bookkeeper's gonna evolve their role to do stuff. So it's it's about everybody having a good think about themselves and going, well, how can I move with technology? Yeah.

You know, we all had to do it at some point in time. Uh, this is gonna show you how old I am. But you know, we first of all would deliver. I mean, I never did this bit.

We would deliver CVs, but we would fax CVs over to customers. Then we learnt how to email stuff over, then we, you know, doing different things and video interviews and whatever. We've moved and evolved. We haven't just gone, oh no, I'm not gonna, I'm not gonna use email.

I'm gonna walk my CVs around. You know, we've learnt to evolve as a society. This is probably where our cohort of of professionals have had to evolve the whole way through our career with technology. This is just another step.

It's a big step, a significant step change, but it's nothing that we can't all work with. 100% and evolve with it. Like you touched on this a little bit at the beginning of the conversation around, you know, how you've embraced AI and how you use AI. And as a business, you've got your sort of corporate license.

So it's a bit more control in a sense. And I love that idea. You know, I've even spoken to some companies, or we've spoken to some companies, about this, where how great would it be in a situation where you've got a high performer or a high value role, they leave or they retire, they take all that IP with them. And as you know, maybe with retirements you can predict it, but with somebody resigning, you know, if they resign on Friday, tomorrow, and decide to go to a competitor, often they they might get walked and that person straight out the door, that IP is completely gone.

So I think with with AI tools now, particularly if you're in a business that's constantly using them, you know, AI should be able to spit out in a matter of seconds what that person did to the decibel on a day-to-day basis, so the next person can come in and plug and play. I think that's pretty cool and fascinating. I think that's really exciting. Again, I think it depends on on the role, but for a lot of roles, it would definitely be that.

Let's think of a role, payroll. Payroll where there are lots of bits and pieces that change in there, and it's not by no means an easy job. And gosh, I love payroll people because they get it right every month. Yeah.

But we would be able to see the specific tasks and when they do them and how they do them every single month. So you're right, that that learning could be within a tool. And I think that you know, we're working with our finance team on how we can make their lives better because we want to ensure that they are also benefiting. I don't think that the evolution to AI can just be in one area in a business.

I think it's you know across the business. So everybody has to be moving together. In regards to the environment that you're working in where, you know, it's consulting, it's client-side, in an AI-led organization as well. Talk to me a little bit about culture.

I mean, how do you build it, maintain it? You know, do you have a certain focus for it? Is it a priority? I have spent a very portion of my career building cultures for organizations, and it is my passion.

Um, I love working with people and building culture. And I think that, you know, when we're looking for people, yes, we do ask them how they're using AI in their job. And I don't think that there are many organizations where they're not asking that question at the moment, but we're looking for people who anyway embrace the future. And really the questions that I'm asking right now are not, do you have these skills or do you have that skill and where did you do it and how long have you had it for?

What I'm asking now is talk to me about the way you learn. Tell me when you've been curious about something and what you've done to educate yourself or solve a problem, how do you approach a challenge? For me, it's all about the problem-solving aspect. And then we're building this community of really curious people.

And part of having a community of really curious people is having a safe space for people to make mistakes and ask questions. And uh whenever you're moving to a new, new technical uh stage, as we are now as a as a wider community, there are going to be mistakes, and there's gonna be heaps of questions. And there's gonna be people discovering new ways of doing things, and that's why we're seeing new tools and new platforms and new evolutions of existing platforms every day coming up.

And some of those things are for me are pretty scary. Are they scary because they're actually scary? Are they scary because I just don't know enough about them and I'm, you know, cautious? Yeah, they are, but you know, people were frightened of cars too.

So I think it's really important that we when I'm building a culture, and the culture that we're building here is really like how good are you and how adapted are you to learning and how curious are you to finding new solutions to problems that we thought were solved. Yeah. I love that. That's great.

I mean, I get excited. I I love working with technologists because they are doing just amazing work all the time. Yeah. And it's just taking us to next level, next level all the time.

And that's just an amazing, amazing space to be in. I'd agree with you. Like we've got an AI partner and they do a lot of the heavy lifting. But you know, whenever we catch up, I I like playing around with the development, you know, I like getting in the trenches a little bit.

And, you know, they're like, oh, you don't need to do that. Like that's that's our job. But I'm like, I I kind of nerd out on this stuff. Like I find it quite interesting.

I'd I'd prefer to know how it's working and and what's making it do what it's doing, so that when we have conversations like this, I can actually explain not just the the outcome, but um how it's actually set up and and what what it's doing for us and and how beyond the same, I I nerd out on this stuff. And it's been really cool to put stuff into play, you know, years ago that has now really sort of evolved. And you know, some of the stuff that we did from an AI perspective four years ago or even five years ago when we started the business, that's redundant now.

You know, we've moved on, we've evolved, we've got different tools. The tool that did one small little thing is now been acquired by a major player. So now we we use that major player's full toolbox of surprises. So it's cool.

Yeah, it's cool. I don't think it's going anywhere. I don't think it's gonna go anywhere. I think it's will become, just like everything else, a part of our lives.

It's not going to be a fad. You know, I was speaking to my mum the other day and we were talking about mobile phones, and she was like, I was certain it was just gonna be a fad. I just didn't think everybody would have one. She says, now if I forget my phone, I have to drive home and get it.

Just as we're sort of closing out, I mean, everything we talked about today, AI, structure, culture, decision making. I'm sure a lot of the HR and talent uh listeners that have tuned in will get a lot of value from today. I mean, if we set down again in two years' time, I'll go, what would you hope HR leaders got right about this whole evolution and shift? I would hope HR leaders really were the leading front on how AI is going to work because we are going to see a lot of roles really change shape.

And if we are not on the front foot, it's gonna be pretty, a pretty difficult road to walk down. So I would hope that we AI leaders are almost sorry, HR leaders are the leaders of AI. Because we will be working in workforces where our our employees will have a number of agents doing tasks for them. And we will need to be able to understand where those what different value add our human employees bring and how they're utilizing the tools around them.

This is just another tool. Yes, it's gonna change some people's worlds, but it's gonna change a whole bunch of other people's worlds for the better. When computers started and people could type faster, it didn't stop, you know, people needing EAs and receptionists and and all those things. We still need everybody.

We just might need everybody in a slightly different way. Yeah, love it, love that. And fingers crossed, we do sit down again and talk about this for two years' time. I think it'd be great.

I would love to talk about this in two years' time, but I think we're even gonna see if we sat down and spoke about this again and approached the same questions again in December, yeah, we would be having a totally different conversation because things are moving really fast and it's really exciting, but we all need to get on the train. I'd agree with that. Yeah. One of the most exciting things I find from a HR perspective is the fact that in decades prior, technology seemed to have a lot of ownership and control from the IT team for obvious reasons.

I feel like AI has now given us this opportunity for HR to actually drive and implement a lot of the change. I've spoken to a number of other chief people officers in my community who own technology now. You know, they're operating in official or unofficial chief people technology officer roles. And the expectation by the businesses, you know, they're they're people experts, but also tech and AI uh fluent as well.

So I think um that's an exciting area. Absolutely. And that's why I really encourage every HR leader to, I know we run hot. I know we are running from pillar to post, and our days are very reactive a lot of the time based on the type of organization that you're in.

But halving out significant time each week to invest in our upskilling and our ability to understand what's happening and where the market's moving using AI and what influence that's gonna have for our organizations is paramount to us doing that important strategic work. The work I like to think of as above the line. You know, it's not the hands-on stuff, it's actually thinking strategically how is this going to affect the business that we're in? How can I add influence to that?

Because we have maybe for the first time ever, the most listened to voice at the table. Yeah, I agree with that. Yeah. And I think it's important for those as well, if they do have the opportunity.

And when I say those, I mean HR leaders to actually own it. You know, if you're in a business right now and maybe they're not as progressive in AI, you know, maybe own that piece, maybe put together a use case for the CEO and own it and you know, take accountability for it. You might have a few fallovers and and speed bumps here and there, but realistically, if you if you do it and you you try it and you experiment, you're gonna see see results. But I know that you are you're having a a busy time this week, Oliver.

So I won't keep you for too much longer because I could, as we both said before we jumped off, we could nerd out about this stuff and talk all day. But I do. It is a problem when you get two talkers on a podcast. They uh they can spill over.

So I would just say, you know, I'm I'm excited about what the future is for HR, and I can't wait to see where we end up. Absolutely. Appreciate your time again. Thank you, Aug.

Thank you, listeners, and we will catch you on the next episode. Thank you. Bye.