Maximizing AI in L&D: Ross Stevenson on adoption, strategy, and staying human

Ross Stevenson [00:00:00]:
All the research that I’ve seen, especially the last kind of three years, is what it all points to is that boring and basic stuff like emails, like responses, summarization, very simple data analysis across the board. Those are in the top five use cases everywhere. So the real ROI at the moment for me is just giving people back time. And that’s a beautiful thing because like I said, we can’t buy time. Time is finite. That’s kind of the big the big promise of AI.

Hannah Beaver [00:00:33]:
You’re listening to How to Make a Leader, a leadership development podcast from Big Think+, where we take the best ideas from the biggest minds in learning and development and distill them into actionable insights. I’m your host Hannah Beaver. Today we’re talking about AI and specifically how to use AI in the context of learning. Now, I’m pretty sure this is not the first podcast episode you will have listened to on this topic, but I’m confident that you’ll want to listen to this one. Today I have the pleasure of speaking with a learning technology expert extraordinaire who lives, breathes and works in all things learning tech. If there’s a tool out there in the market that you’re curious about, then chances are he’s likely tried it. Ross Stevenson has spent nearly two decades in the L&D learning technology space, working for corporations such as Trainline and Tesco, and currently as the Chief Learning Strategist with Steal these Thoughts!, his company, where he advises L and D teams on how to improve their performance with technology and AI. I’m a big fan of his newsletter, which is packed full of insights on the latest and greatest in learning tech and tools, and also plenty of memes.

Hannah Beaver [00:01:45]:
In today’s conversation, Ross will share his thoughts on the best ways to be thinking about and using AI tools in the context of learning. He’ll be sharing some tips on the best ways to use ChatGPT, as well as some of his favorite tools in the space right now. Let’s get straight into it. Ross, so great to have you on the podcast today. You are somewhat of a learning technology wizard, so I would love to hear how you got here and if you could walk us through the last 10 or so years of your career.

Ross Stevenson [00:02:19]:
Yeah, well firstly, thank you for having me. Lovely to have a conversation. And yeah, it sounds like my superhero origin story, which I’m probably going to remember very differently every time someone asked me for it. So the tldr, as I say to a lot of people, is I have been in the HR L&D biz for about 17 years. Never planned to be here. My whole education is in technology. I was supposed to be a software engineer, building products, doing all that stuff over in Silicon Valley. That didn’t happen.

Ross Stevenson [00:02:48]:
I kind of fed into this industry and it was kind of a strange marriage of, I think very quickly I understood there was a real lack of digital technology understanding in this industry. Digital learning wasn’t really a big thing kind of 12 years ago when I kind of first dropped into it, you know, now it has been all the range and the last 10 years in particular have really just been focused on L&D from a technology perspective. So what I mean by that is I’ve had really odd role titles and roles that you won’t consider traditional L and D, but fundamentally what they focus on is saying, how do we transform what we know as workplace L and D, which for most people is this kind of really strange quasi education model that kind of doesn’t really work in the workplace because of how we live and how we work. And how do we start to use different types of technologies, whether that’s platforms, whether that is, different tools that we could bring in to actually help people move more towards performance. So I know we talk about learning a lot as an industry, but the reality is when you’re working for big, you know, 5,500 companies, the end goal is performance. It is the bottom line. It is getting that roi. So a lot of my time was spent on building digital systems, building products that were able to let people get the knowledge they needed to perform.

Ross Stevenson [00:04:11]:
Nowadays, I’m on my own, so I’m solo, I do my own thing. I think the popular term is solo founder that I see on on LinkedIn. So I have my own business, I do my own thing. Basically what that means is, for those of you who may know already, I have a newsletter about learning tech that I’ve had for years that thankfully people subscribe to, they engage with, they pay me for, which is great. So in a weird way, it sounds odd for my job today, but is fundamentally taking all those user knowledge, helping people to understand how they can apply that. And then I also work with companies as well, specifically on We’ve got tech. How do we acquire tech? What’s the right tech? How do we use that intelligently? Obviously intelligently being the key word there, because anyone can buy tech, but actually buying the right stuff and use it intelligently is the biggest thing. So I hit TLDR, but it wasn’t very much TLDR, but basically 10 years learning tech, making it work for people and basically just helping people unlock the power of it.

Hannah Beaver [00:05:09]:
I love that and thank you for sharing. I love the full story. I do love that you mentioned you never really intended to end up specifically in the L and D industry. This is a conversation that I’m having over and over with multiple guests. It’s always interesting to hear how people got here. I love how you talked about using technology for connecting people and connectedness. Especially, as you said, because technology is so overwhelming right now. I think it can make people feel isolated.

Hannah Beaver [00:05:33]:
Know we’ve had a podcast conversation about multiple generations in the workplace as well. I think that could have a part to play as well. So I think just grasping that understanding of technology and how we can use it, you know, harness the power of technology to make ourselves better at our jobs and also to connect each other, is something that I love that you just mentioned. What is an opinion or perhaps a hot take that you hold around learning technology that is maybe a little bit controversial?

Ross Stevenson [00:06:01]:
Oi, I don’t know if it’s controversial. I just think much to kind of what I was saying earlier on, I just think there’s a surreal misunderstanding and underuse of what I would call intelligent digital technology use. I think we have so many apps, we have so many different bits of technology, but people barely scratch the surface. It still astounds me all these years later where I’ll go into companies and they have really powerful tools like a Microsoft Teams or a Slack, but they just use it to message each other. They have no idea about the capabilities beyond that. And I think for all of the technology that’s available, all of the money that is spent in the industry from a tech perspective, L&D is worth about 400 billion. I think the last time I saw it is that I met very few people that really understand how to leverage it to help them. I find more people that cause themselves problems with the tech and blame the tech rather than trying to spend the time to understand the tech.

Ross Stevenson [00:07:06]:
So I still think, and that for me really hasn’t changed over the 10 years, is that what I class as digital intelligence is severely lacking in the HR&L and D world. And the answer to that would be, I think a lot of people have always gone, well, tech’s not my thing. I don’t think you can say that anymore because the world is run on digital tech. Just saying tech is not my thing or I can’t use tech is. It’s not the way forward, unfortunately, as brutal as that is. So.

Hannah Beaver [00:07:34]:
Well, that’s why I’m very excited to have you on today. I personally have been a subscriber to Ross’s newsletter for a while and I know on LinkedIn you share really great informational videos as well. Kind of walking through new technologies that you’re evaluating and using and kind of sharing stories around how best to apply them in L and D setting. So I think it’s going to be a great conversation today and I’m excited. What is a common misconception about AI and L and D that you’d maybe like to clear up?

Ross Stevenson [00:08:05]:
Oh my. How much time do we have? I mean, that’s. I mean, there’s many things. There’s many things. I mean, I really think we’re in this, I don’t know, matrix, like environment at the moment, where I feel like everyone and their dog is an AI strategist at the moment, like trying to give me some kind of expertise or something like that. I think the thing for me that really stands out is to kind of clear up is that it’s much more than tools. I think people really get themselves lost in this. Oh my God, it’s like a million AI tools and I have to know how to use this, this and this.

Ross Stevenson [00:08:37]:
And the center of it for me is really about it’s a new framework really for the kind of next stage of whatever you want to call this error is that we’ve kind of had this one framework for the last, let’s say 10, 15 years, which is Web2, where you can read information, you can build your own information online. And then, you know, some super nerds won’t call this web free, but if we class it as the kind of web free version, is that now you’re able to have context and have conversations with information and data. And I think in that is there is a philosophy of actually understanding how does that stuff work and why is it useful. We’re getting so excited about all of this stuff. We don’t really ask the questions of, well, what could be unintended consequences, like where could this go? What can we learn from other stuff? So I think the biggest misconception is it’s just thinking it’s tools. It’s just thinking like, I’ll go off and I’ll learn how to use ChatGPT and I can use AI is mad, really, in my opinion.

Hannah Beaver [00:09:41]:
It’s like the shiny new toy, you know, that’s in the room at Christmas time and you just want to play and you’re neglecting, you know, the trampoline in your backyard that is you know that you’ve been for years and works well. So with these new AI tools and like you said, people are kind of seeing the tools. I’ve got to perfect these new tools and then I’m good to go and I’m going to be, you know, advancing ahead of my career. What would you say is the biggest mindset shift that L and D pros should think about as they start integrating AI, AI and new learning technology into their work?

Ross Stevenson [00:10:14]:
Yeah, I think the biggest mindset shift is more going to be around why are we using this stuff like anything at L&D in most industries is kind of still being product minded and saying, you know, what are the problems we’re actually trying to solve? And you know, I make fun of this online a lot and people who see my stuff will see that I probably share way too many memes poking fun at this that I have conversation with companies is, yeah, they, there’s too much of just chasing tools. It’s just like, oh my God, I’ve seen this new tool and I’ll use our tool now. But there’s so many companies that I work with that they just don’t know the problems. They don’t know the problems they’re trying to fix. They basically got a tool and they’re trying to find a problem for it, which is the worst case scenario to be in. So the mindset shift for me is understanding that this whole AI era right now is actually more about behavior change. It’s more around I mentioned before about the framework of if we kind of imagine or even imagine, we look at the last 20, 25 years, the way we’ve kind of looked at the world is through Google search. So it’s become a verb, Google.

Ross Stevenson [00:11:17]:
We say we’re going to Google it, we’re going to find information, we’re going to do that. And now in the last few years we kind of move to this framework of oh my God, I’m going to converse with information now, I’m going to ask questions, we’re going to have a natural conversation. That is a huge shift and we’re having at work as well where it’s not just saying I’m going to go off or build a course on my own or researchers on this own, I’m going to have a conversation with an entity that allows me to kind of draw these insights that I wouldn’t have been able to have done 18 to 24 months ago. So there’s a far bigger behavioral shift in understanding that this is a new way of working, a new Framework of how do we create information, how do we gather information, how do we assimilate information, and then how do we put that into action? But like most things in life is, I don’t think that’s really seen yet. For a lot of companies, what they see are tools. What they don’t understand is if you give people a bunch of tools and you don’t give them the right behaviors and take them on the journey, you will suffer for that later on. So it may feel like now and you see the headlines like, you know, AI saves me X amount of time or it’s this productive. It’s like, it’s really interesting to me.

Ross Stevenson [00:12:32]:
How do they measure that? Like, how do they measure productivity? How are they measuring all those things? So I think that the biggest mindset shift is just getting around at the fact that this is a real behavior change and behavior change takes a lot of time and companies want this change to be in a couple of weeks because they see all the big headlines of AI, AI, AI. But actually, you know, to really kind of take advantage of it, you’re probably looking at years. It’s not a sexy thing to say, but in reality, if you’re looking to get the most out of it and you’re looking to say, how does my company move from a multimillion dollar company to a billion dollar company? You know, that’s a probably multiple year, multiple decade journey. I think that’s really the same for these generative AI tools at the moment, because a lot of them are still kind of very gimmicky, they’re kind of fun. But you still see for most people, most organizations, it’s really hard for them to kind of place and say, well, is this actually helping us or is this just kind of the cool thing right now, but can it help us, you know, many, many years down the line? So yeah, the biggest mindset shift is definitely looking at the treat it as more of a behavior change project rather than saying, this is just a training project where we’re going to get a tool, we’re going to train people on this for two weeks and then they’re going to go off and they’re going to be superstars and the company is going to be very productive. That’s 100% not going to happen. And I think a lot of that is false flags. And, and I see data online already where I’ll see people saying, yes, you know, we’ve had a thousand people use this tool.

Ross Stevenson [00:14:06]:
That’s not really a metric that you want to invest in. The metric you want to invest in is saying, you know, ChatGPT saved me $10 million in two years. Those are the kind of things you want to look at that we don’t talk about. So yeah, in sum, look at it as a behavior and change project that you’re part of versus saying we’re just going to train people on a new tool because they are new tools, but they’re part of a newer technology that the world is now experiencing. So there’s a difference there as opposed to what we’re currently used to.

Hannah Beaver [00:14:37]:
I think it’s interesting the point you make around it being kind of a longer term project and mindset shift. I think I’m guilty of the shiny new toy syndrome almost. I know when in our prior conversation I had asked you and proposed a question today around the idea of give me a case study who’s using AI tool, who’s doing it well, and you kind of responded to me that we don’t really have that yet because it is such a work in progress. So I do think it’s interesting that we have these new tools and we think that they’re going to kind of work immediately. It’s this silver bullet that fixes all and streamlines all processes and improves, you know, the learning process. So I think I’m guilty of that mindset too.

Ross Stevenson [00:15:18]:
Yeah. And it’s really hard, right, because look, ROI is different for everyone in terms of their context. So. And I say this to people all the time. It’s like for some companies, having a tool that helps people craft better emails could be hugely beneficial for them or helps them craft better copy in what they’re working on as a marketing team be hugely beneficial for them. But that same context based on a company of 10,000 people, that’s FTSE 500, that works in the financial industry, that’s not going to help them. So their use case and their context and their ROI is different. So the thing I look for, particularly in case studies is saying it’s looking at something incredibly specific in that’s had that marginal gain.

Ross Stevenson [00:16:02]:
And I think you’ll find lots of stuff at the moment, especially in the L and D space where it’s like content, content, content and there’s little things where it’s like, you know, review my content and do this. They’re great, you know, contextually for that. But I mean when we talk about case studies, it’s kind of like I’m looking at like a huge company that’s kind of done the last five years we looked at this and these are the things that we’ve improved and this is the how tangible AI was in that. I, from what I see in my own research, I just, I don’t think there is anything that I would look at and say is kind of like high level right now. But then again, I don’t expect that because the technology is so new. It’s kind of one of those things where people hate when I say it’s like, yeah, well, come and talk in 10 years time. A lot of technologies now we can talk about at this moment because they’ve been around for so long and we can see the success. Like, you know, you got only got to look at streaming.

Ross Stevenson [00:16:55]:
Like everyone laughed at streaming when it first came out. It was all Netflix. Who, like, who are these people? They’re not going to do this. And now it’s the biggest thing, but you can’t see that at the time. It kind of takes all that time to say, oh my God, you know, this works. And it, it works really well. And these are the case studies to, to say that. But it’s really all based on people’s point of view.

Ross Stevenson [00:17:14]:
Your context is again, very personal in how does this help you? And I know people that feel like they have been transformed in some ways from using these tools, which is great, but it’s very personal on that as opposed to, I have my business hat on and I think about ROI and performance increases in terms of actual monetary gain.

Hannah Beaver [00:17:36]:
Curious if from people that you’ve worked with, if you do have any anecdotal stories or just examples of how a client or a company you worked with has used an AI tool effectively. And maybe, you know, we’re not seeing the ROI yet, but what are some of the positive impacts that you’ve seen or an example of how that’s really been put in practice?

Ross Stevenson [00:17:56]:
Yeah, I suppose this couple, I mean, some of them are very, what I call boring and basic. However, I think boring and basic is, is probably very hot and very sexy right now because, let’s be frank, we’d rather I do the stuff we don’t really want to do and to allow us to do the exciting stuff rather than for whatever reason people seem to want AI to do the exciting stuff like the writing and the building and then what gets there for us. Whereas I’m all like, no, get it to automate, you know, responses for lead generation or get it to look at how is it going to improve your process in creating user stories as a product team or a tech team, or answering support queries. Get it to do those things. So then you can go off and do those other components. So in terms of specifics, I think this one and I can’t name the client, but we’re just talking about there. So working with their product team to look at, and most product teams, I think product people listen in between product and tech teams, you always create in these user stories, which is basically like your fictional users and how they use the product and what they want to do. Now for these product teams, they can build hundreds of these especially.

Ross Stevenson [00:18:57]:
They’re a huge organization with huge products and that can be incredibly time consuming to do. And you’ve got to take a bunch of notes from interviews and whatnot and structure them and turn them into this. For me, that is something that’s kind of pretty ripe for these generative AI tools at the moment. Because it’s one specific task. You can template that and you can just basically say, here’s a template of what good looks like. And then what we’re going to do is basically give you all our notes. This is what good looks like. This is the template.

Ross Stevenson [00:19:27]:
Create this for us. And then you can create these user stories, which sounds really small, but in that circumstance, that team was able to save like eight hours over the course of a week as a team. So it doesn’t sound huge. But for them that’s like, wow, you know, time is money. You can get more money, you can’t buy more time. You know, another one with a different tech team where looking at ticketing support. So most companies have a ticketing support system where a user will raise a ticket to say, oh, I’m locked out on my laptop, or I can’t use the printer on site. And usually there’s like a system somewhere that kind of helps filter that.

Ross Stevenson [00:20:04]:
It gets scheduled over to an engineer. You’ve got to wait to the engineer to come back. But you can put a first layer of an AI assistant in that that basically just says, well, am I able to deal with this query based on our knowledge base? What I’ve discovered, which is quite interesting, is that most of the time a lot of the queries that come into these people can be solved by the user. They’re really simple things like password reset about using a certain application or requesting a tool. You know, I think in one case there was like something like 65% of requests could have just been answered by the user. It was just them being lazy and not looking for the information.

Hannah Beaver [00:20:39]:
Your computer. I know, that’s exactly, yeah. Have you pressed the on and off button 100%.

Ross Stevenson [00:20:46]:
Yeah. So I think time is a very interesting metric in all of this stuff because people look for like the real grandiose use cases like they want to see, you know, one person businesses and all this rubbish and AI running everything. I’m not really impressed by that. I’m more impressed by, well, we’re already overloaded. How do we kind of help us take some of the stuff away? That’s so repetitive, really kind of bogs us down and allow us to do the human stuff. Companies I worked with where we’ve looked at doing stuff for performance reviews and performance reviews bots that helps coach managers and gives them resources to help prepare them for performance reviews and then helps them hopefully anyway give a better quality performance review experience. Same with onboarding as well. So onboarding at a lot of companies can be really choppy, but for small companies they kind of really struggle because there’s usually not a lot of resource to focus on onboarding.

Ross Stevenson [00:21:38]:
But you can introduce kind of AI enabled components like an AI buddy that basically helps new starters and managers alongside those processes in terms of getting them the right information at the right time, getting questions, answering quick things, you know, just real low hanging fruit. I think one more bigger one and it’s not one of mine, it’s a public one from an Australian mining company called BHP Mining, where I was reading something one of their team put out about. They were looking at their leadership framework and it’s a couple of years ago and they were basically saying, you know, they pay McKinsey, it was 1.6 million Aussie dollars to come and do a leadership framework for them. And they were like, well, why don’t we just stick it into ChatGPT and say ChatGPT, this is our framework, this is what we’re trying to achieve, this is where we are today and just use it as a thought partner. And that team, in that context they were saying not everything was great, but they were really impressed by most of the stuff it came out with. And they were looking at and saying, well, you know what, actually we could take a lock from here and then take that into other humans and then use this to start doing stuff ourself. And what was interesting about that is that something that would have cost a company $1.6 million to go and get an outsourced company and there’s no guarantee that actually they’re going to get what they want. They could actually bring that in house and use tools to say, well, how do we keep this close to us and what we’re doing and have more control over that and give us different points of view and help us build structures and assimilate best practices from other companies to do that.

Ross Stevenson [00:23:18]:
All the research that I’ve seen, especially the last kind of three years is what it all points to is that boring and basic stuff like emails, like responses, summarization, very simple data analysis across the board. You know, you go to Gallup, McKinsey, BCG, Harvard, all of reports say the same thing. Those are in the top five use cases everywhere. And that makes sense because if we get that stuff done and we can do it to a decent level, then we can go do the human stuff, which is more like, I’ve got more time now to actually go speak to a stakeholder and really understand their pain points and have a conversation and build a better product instead of being drowned in 400 emails from the business around all of these components. So I think the real ROI at the moment for me in these use cases and kind of little mini case studies is just giving people back time. And that’s a beautiful thing because like I said, we can’t buy time. Time is finite. So for me anyway, that’s kind of the big, the big promise of AI.

Ross Stevenson [00:24:21]:
I’m not so concerned about other components of automating my life and being boring as hell and just sitting around and watching the world go by.

Hannah Beaver [00:24:34]:
Chat GPT, we’ve already referenced a couple of times that we definitely cannot have this conversation without digging a little bit deeper into ChatGPT. It’s a tool that everyone has access to and everyone has used, I’m sure, in some element in both their professional and working life.

Ross Stevenson [00:24:53]:
Yeah.

Hannah Beaver [00:24:53]:
So question for you to get a little bit more into the nitty gritty. How should L and D Pros be using ChatGPT effectively? And what is some of your guidance on best tips and tricks to effectively prompt ChatGPT?

Ross Stevenson [00:25:07]:
My first controversial point would be do you need to use ChatGPT? I would say before any tool, go figure out your problem. So I have to say that and doing what I do, like define your problem and then figure out if AI is a tool that can help you with that. If you do decide that and you go down that road and it is chatgpt. So I think this is an L and D perspective. What’s really obvious is content creation is the hot, sexy thing for people in L and D with any of these tools. For me, that’s kind of like 2% of its power. And unless you really understand how to prompt it and You’ve really got kind of a clear mind of what you want to achieve, what success looks like, you’re not going to get a good result. And I think we’ve all seen stuff online where people like, I can create a course in two minutes and blah, blah, blah, like, yeah, but let’s be honest, 99.5% of them are utter crap because no one puts any thinking into that.

Ross Stevenson [00:26:02]:
So my first thing would be just kind of move away from content creation a little bit and then I will look at other components of. So in my own practice, what I found very beneficial is how do I use it as a thought partner or a critical thinking partner. So what I mean by that is, you know, I write a lot. I look at a newsletter as a product. I put a product every week that goes out to like four and a half, 5,000 people. I spend 20 hours on it a week. I am lost in my own mind half the time. So to get a critical view from something else that is not me or a devil’s advocate, or to challenge my ideas or just say, are these 3,000 words just a bunch of rubbish that I’ve done by drinking too much tea? Or is there something in there that’s going to help someone like that is profoundly helpful.

Ross Stevenson [00:26:49]:
And I look at that for many different types of work. So companies that I’ve worked with where people in sales teams as an example when they’re responding to proposals, and I’m saying, well, if you’re not too sure how to respond to an email, drop it into ChatGPT and basically, look, I’ve got this email, here’s a scenario, here’s what I think. You know, give me Devil’s advocate. What are the unintended consequences if I say X, what am I missing here? So this is beyond using this tool for creation. It’s all about challenging you. It’s hit and miss sometimes, I must say, sometimes you’ll look at it and it’s just like, oh, God, that’s, that’s rubbish. But sometimes if you structure what you’re saying, and that’s the real key thing here, you still need to have really good clarity of thought and structure your writing. You can call it a design partner as an L&D and say, you know, I’m building this workshop or I’m working on this program, here’s what I’m considering doing.

Ross Stevenson [00:27:41]:
What else could I consider? You know, these are the challenges or these are the pain points. There’s just a lot of use in there for me with that. And I think for me personally, I use it a lot for that, just as a sounding board. The other two I would say is research assistant. So I say chatgpt slash perplexity for this. I think perplexity is probably better than ChatGPT in that domain. And research as in so there’s two folds, so there’s a very simple stuff of Here’s a PDF, summarize that for me. That’s cool.

Ross Stevenson [00:28:10]:
But I think the other end of it is that, you know, let’s say I do a lot of work in when I’m doing keynote speeches, I’m basically given a thesis on a topic and what I want to understand is what the pros and cons of that thesis. So what already exists in the world to do that. And I think as a research assistant, again, I found these tools really, really useful. So when I’m going to give in these keynotes that I’m going to have a balanced view of where might I have my own biases and where’s the stuff that kind of makes sense and data agrees with or you know, what am I dubiously speculating on and don’t have data kind of to actually back up with and using it as a research assistant to compile, that is, you know, again, I’m just amazed by that consistently in terms of speed, quality, structure and actually how that helps me build better products in what I’m doing. And then the final one would be around data analysis. Like most people that operate in their own businesses, I have loads of data touch points with that’s from financials, engagement, all of these kind of different things. The ability for me to take any kind of type of surveys or CSV files or whatever and drop that into these tools and have a conversation and just say, you know, well, I’m looking to do this. What can you see here in terms of a pattern or I want to get a sentiment analysis on a bunch of data that I’ve just taken from a thousand people.

Ross Stevenson [00:29:30]:
You know, help me get to that quickly. That saves me time, I think for partnering research assistant data analysis and then my bonus one would be definitely using it as a skills coach, as a foundational level to try and kind of help you role plays. But role play of ChatGPT in terms of I’m trying to learn this new skill, I want to practice this skill, help me do that.

Hannah Beaver [00:29:53]:
I really like the multiple Personas that you presented there for ChatGPT, you know, moving away from content creation, but a thought partner, a research assistant, a data analyst, just like an interesting way of thinking about it. So thank you for sharing. So I have noticed from some of your work, I love the quote where you mentioned that. But the current state of AI use for work is like giving a Ferrari to a five year old. So where some people don’t have the skills, experience, or know how to use it effectively. So what are some of the biggest mistakes that L and D professionals specifically are making when it comes to using these new AI and learning technology tools?

Ross Stevenson [00:30:36]:
Yeah, the biggest thing for me is not understanding the fundamentals. So what I mean by that is that we spoke earlier on that I look at these as more than just tools. What we’ve basically introduced into our framework as a new technology. Now, yes, I am a tech geek, but I think the best way to leverage and maximize tools from a technology stack is to understand at a very base level how does it work. I think this is even more important with generative AI because we have this caveat of hallucinations where these tools can make up stuff because they’re basically looking at patterns and they’re creating their content from that. And we have to just recognize that it will lie. That’s part of the package. It’s a feature, not a bug.

Ross Stevenson [00:31:23]:
Until these tools improve, too many companies ignore that. And actually you should embrace that. You should understand that that’s going to improve your use of it. And the way it improves your use is understanding when it’s the right time to use it for a project and when it’s not to look at the sensitivity level. And then for the user, how do they, not just at work, but in life, how do they create the behaviors and skills that helps them spot stuff? Like, you know, when we talk about misinformation or fake news or any of that stuff, I mean, you know, they were recording this, it’s election day. As we’ve spoken about in America, I’ve never known a time over the last few years where, you know, misinformation, fake news, generated pictures, videos from AI have been so big. And we’re going to have a struggle in organizations, in people actually understanding that when is the right time to use this stuff based upon limitations? We just have to recognize that in all things in life there is risks and there is limitations, but that comes down to just getting clear on the fundamentals. And the reason why I say that is I work with too many teams where the first question is, how do I use this tool? I’m like, no, no, no, we need to go back and think about how do you understand what generic AI is? Then Say, okay, I can then go and make the right decisions.

Ross Stevenson [00:32:41]:
So when my company does give us AI tools, I can be clear on, here’s where it’s going to be really helpful for me. Here’s where it could be harmful, but here is how as I as L and D Pro can make smart decisions on outputs. How do I evaluate them, how do I challenge them, how do I have watch outs for. And I don’t think that’s a big task. Like the fundamentals of that I’m literally talking about you could go and watch a half hour YouTube video and you’re sorted as a beginner, as an everyday human with this stuff. But my personal opinion is that will 10x your output with performance in these tools. And I see too many people making mistakes because they don’t have that knowledge. So they’ll make mistakes in the potential of the tools, they’ll make mistakes in how they’re deploying tools.

Ross Stevenson [00:33:28]:
And I think in the, the L and D and HR arena we deal with too much sensitive topics for that to happen. We really need to have people who have a clearer understanding on, yeah, it might be great to build a conversational bot, but actually is that going to be the right thing for us to do? Is that going to help us deliver on what we want to do or are there going to be risks there that we can’t take on due to the sensitivity of that information? So for me it’s really just getting those fundamentals. If you can get the fundamentals, it will help you out so much.

Hannah Beaver [00:34:11]:
Do you have any, any other guidance or maybe like a framework with teams that you’ve worked with to ensure that this, you know, for example, we have a team that doesn’t really use any AI tools or learning technology tools. They want to introduce new tools. How would you ensure that that experience is positive and not disruptive?

Ross Stevenson [00:34:29]:
So my view would be it’s really wider than the L and D teams. And I think this probably, and people listen to this probably experience all the time, where very usually, and I see this for my last 20 years, is that you’re at a company, something new happens, a new bit of tech happens and the company just decides that the L and D team needs to make everyone use it and that the L and D team on its own is going to change all of these pre built behaviors, they’re going to change the culture and they’re going to embed this tool and everything’s going to happen. Unfortunately, the reality is that’s not true. It’s a whole organizational Effort. And what happens is that to get the right new behaviors, you need to have that right environment for it. And LOD plays a part, obviously. They are a mechanism in delivering to people and helping people with that. But going back to our earlier point, there is a huge behavioral and mindset shift in this technology and these tools.

Ross Stevenson [00:35:25]:
So my advice would I say it’s actually be twofold. It would be looking at you getting clear on the fundamentals, but also making sure you’re connected into the wider organization in how are we as an organization approaching this as a change project, so together we can make this successful. Because my hot take would really be you’re not going to be successful as NND on your own, like just going out there, whatever you want to do, put out a bunch of training programs or upskilling programs on this one tool. It’s not going to make people as confident or as competent as you might think it will with you alone. But you, in conjunction with senior leaders, with tech teams, with all the different teams that are part of that environment that you’re trying to change, can definitely make a difference. But I think L&D as the mechanism alone, it’s really difficult. And it’s really difficult because outside of this conversation, but you know, most companies look at L and D teams as a nice to have rather than like a legit serious function that contributes to business change in that. So we’re also dealing with that in terms of huge transformational technology being delivered by a space in a company that people don’t really give the credit that the industry believes it should do and then wanting that change to happen in.

Ross Stevenson [00:36:48]:
And I always say, you know, senior execs always want you to do stuff in a week that needs a year to actually happen. So it’s really difficult for that. So, yes, you can help yourself get clear on the fundamentals. But like any good L and D team, in my opinion should do is about how do you connect with other parts of your organization, how do you build relationships. So when you do go do this, you know, adoption program with training or whatnot, with that one team who doesn’t really seem open to it. It doesn’t just end there. If you just go and do, okay, we’re going to do a half day tool training, that’s not really going to change anything like it might. Some people might go there, some nice ideas, but very few people act upon what they actually experience in those sessions.

Ross Stevenson [00:37:32]:
So how do you make sure that that actually is embedded by working with these people and looking at this whole thing as more of a change transformation program, which for a lot of companies, actually big companies, you know, changing transformation programs are three to five years. And I appreciate that is really, really difficult. That is not going to be an easy ask. But my whole kind of focus, this message is that don’t feel like an L and D. It lives and dies around AI adoption being successful because of you, because it really doesn’t. It’s also down to many other people in the organization being part of that process. And what I would say is the, the companies that do well with this stuff are that they adopt it all across the field. So one example of that would be, and you can find it online on OpenAI’s website, is they did a case study with Moderna who obviously were one of the organizations that worked on the COVID vaccine and you know, they had ChatGPT for a year, but they got the whole company involved in terms of behavior change, in terms of training, in terms of embedding that, you know, everyone was bought into it and they’ve had the success that they’ve shown in that customer story because everyone was part of that journey.

Ross Stevenson [00:38:49]:
It wasn’t just OpenAI worked with the L and D team to try and convince the rest of the business to do it. The rest of the business was on board and doing that. So I would certainly say, look, in L and D, what we’re responsible for is of course helping people improve their skills, improve performance. You’re not going to be the singular agent of change in terms of bringing people into, you know, the AI world. And I think if you go down that route, you’re not, you’re not going to have a lot of success. I think focus on the things that you do really well. Partner with other people to strengthen those areas where you can say, how can we help people adopt this?

Hannah Beaver [00:39:29]:
I think that goes really nicely with your thesis that the future is human powered and not AI first. So on that topic, would love to hear what are the top three human skills that you think are essential to develop in such an AI forward workplace?

Ross Stevenson [00:39:46]:
Yeah, so the main things for me and some of these will cross over. So you’re going to have critical thinking for one. I just, you know, the biggest thing for me is, and I, I disagree with a lot of the stuff online that I see, but it’s AI first mentality and you know, getting AI to this. Don’t outsource your thinking. Thinking makes you human. Like it’s, it is your tool that’s incredibly powerful. So critical thinking is 100% there and alongside that as well is so I mean people call this problem solving. You can was problem solving or kind of like judgment or decision making there is that.

Ross Stevenson [00:40:20]:
My personal opinion is what makes us really good in the workforce. As humans you can help use AI to help you with this is that is solving problems, is deconstructing problems, understanding how do I fix this thing and doing that at every stage. And like I said, critical thinking and this problem solving notion, they do overlap with each other, but there’s still different skills where it’s like you could be a critical thinker but you can suck at solving problems, right? You can think all you want, but if you can’t actually figure out how to solve the problem and construct that, you know you’re not really going to go far on that element. And the other side of it, and I suppose this is more like the gun to my head looking the kind of free. Three things is I’m going to cheat here a bit. Is that so? I had an article that I written where I spoke about this concept of social influence, which I’m cheating here because you can basically break this down into sub skills of like communication and emotional intelligence and stuff. And that is I say social influence because like as humans, the way we work is we’re a social society. Like we tell stories to each other.

Ross Stevenson [00:41:18]:
That’s how we buy into ideas, it’s how we share knowledge in that that is not going to change with AI at all. We’re still going to tell stories. We still need people to buy into our ideas, to buy into who we are, to be seen and valued. Like no AI tool is going to change that for you. Like you’re going to get ChatGPT out on your phone to then give yourself some kind of like social credibility in a meeting to talk about who you are and what you do. You’re going to need to be able to do that. Social influence is a real big part of life because although people don’t like it, you know, you do it all the time. Like you’re trying to show people your credibility in industry, you’re marketing to people about who you are and your skills.

Ross Stevenson [00:41:56]:
Like in an organization, you’re doing that when you want a promotion. You’re trying to sell people into who you are and what you’re doing. And so, you know, critical thinking, problem solving and that social influence bit are incredibly beneficial in terms of you just being able to operate. So which is why I kind of do venomously disagree with the AI. First thing that I see with people, because if you go down that route, what happens when everyone else goes down there? You’re just going to be like everyone else. So you’re just going to be very much in the same field. And the people in my opinion who I work with, different companies, especially at exec level, are they’re great critical thinkers, they’re great problem solvers and they are great in social influence. And I don’t think that will change.

Hannah Beaver [00:42:43]:
This may be a difficult question because I know you are evaluating and looking at learning technologies all day, every day. You had to pick a top three favorite learning technology tools right now. What would they be?

Ross Stevenson [00:43:00]:
So I should preface that. I think any technology can be a learning technology. So I don’t look at these specifically in the industry. I cannot not say chatgpt. I just think, you know, it’s good timing for OpenAI, but they were really the ones that kind of however you want to look to, it got to market first. And the other one right now that I find a lot of benefit in, I think it’s great for work and for research is a tool called Notebook LM by Google, which is really just built around, I can’t even describe it. Like there’s a AI souped up version of notion where it’s all focused on data analysis and reports and research and collating notes and looking at thesis in that. So I think just across the board that’s incredibly beneficial.

Ross Stevenson [00:43:42]:
Obviously Google was huge, they got trillions of dollars, they’re going to keep building those things. So that’s a pretty safe bet in what I look at as a great product versus just a feature. And I think there’s a lot of AI tools out there that are features, I think more from the workplace learning space, if you want to, you know, get specific. I suppose a company that I, I talk to a lot is Sana. So Sana’s a Swedish organization. They’re in the learning tech space, but they’re in the AI space as well. So they’ve also got their own LLM too. And they’ve got Sana AI.

Ross Stevenson [00:44:16]:
And I think they are on my radar anyway. They’re the one of the first companies that’s really come to market in that space to say how do we do all this stuff, but how do we do it specifically for work? So it’s secure. So how do we create assistance? How do we create workflows? How to make it really simple for people to use this at work? Because the two I mentioned previously, there’s always the call back to oh, it’s going to steal our data. Oh, it’s not restricted. Oh, how do we do this? So Santa is interesting to me because again, going back to OpenAI ChatGPT, they seem to be one of the first ones to market to say, well, we’ve got an AI tool that is specifically built for the workforce that does those things, does those things in a secure manner and gives you a easy way to do it.

Hannah Beaver [00:45:00]:
I’ll definitely have to check out Notebook lm. I’ve seen a lot about it and I’m a fan of Notion, so maybe I’ll have to check it out to elevate my data and organizational game. Well, with that, we’ll wrap up today’s conversation. Ross, thank you very much for your time and insights today. I’m sure our listeners will learn a lot and just always fascinating to talk to someone who’s so deeply entrenched and aware of all goings on around learning and technology. So we really, really, really appreciate your insights today.

Ross Stevenson [00:45:30]:
Well, great. Well, thank you for having me. It was lovely to chat to you.

Hannah Beaver [00:45:33]:
For more from Ross Stevenson, check out his newsletter and website, Steal These Thoughts!, as well as his LinkedIn, which we’ll put in the show notes for more from How to Make a Leader. Make sure you subscribe so you never miss an episode. We’ll be back next month and every month with more insights from another L&D expert. Thanks for listening and we’ll catch you next time as we learn How to Make a Leader.

Maximizing AI in L&D: Ross Stevenson on adoption, strategy, and staying human

Maximizing AI in L&D: Ross Stevenson on adoption, strategy, and staying human

Forget the AI buzzwords. Here’s how L&D teams can actually use AI to save time, solve real problems, and keep learning human.

Request a free demo and gain access to the entire learning platform