How to Introduce AI To Your Team
By VICKY BROWN
The world of Artificial Intelligence is exploding every day, new features, new adaptations, new capabilities. By the time this episode comes out, it’ll probably be outdated.
So, all the more reason we, as leaders, need to face this brave new world head on. No hiding, and thinking (or hoping) it will go away, or not impact your business. It’s here to stay, it’s growing, and every business is going to be impacted. It’s up to us if that’s a good or bad thing.
Now, I’m not going to even try to delve into the specifics of what AI can do – there are all sorts of experts deep diving into that all over the place. But we are going to take a look at some of the things you need to think about as your team takes on AI.
What, you thought that no one on your team was using AI? Let me tell you, just because they haven’t come right out and said so, doesn’t mean they aren’t using it left and right. It’s just that you and I missed the memo.
So, if you want to be sure you’re leading the charge, instead of hanging on by your fingernails for dear life – let’s talk about introducing AI to your team.
First up, let’s set some base knowledge. The most popular AI tools are Generative AI or Large Language Models like ChatGPT. Generative AI is a program that can create new things, like pictures, stories, music or even computer code. While large language models are a type of Generative AI that is trained on huge amounts of information and data. They are primarily used for text based tasks such as translation, summarization, text completion etc.
The main thing you need to know is that together (and most of the popular tools today use them together) they can take on a lot, and I do mean a lot, of tasks.
Now, on a personal note – so far AI seems to be most effective when you use it as a research assistant, or brainstorming partner. Just taking AI output, in it’s raw form, and dumping it into a final project or presentation generally doesn’t yield great results.
So, for that reason – as well as the fact that AI still can have a ‘hallucination’ issue – meaning it can just make stuff up – you’ll want to be sure human oversight still carries the day. And while AI has immense potential to help streamline processes, and making things easier all around – it’s critical that you have an eye on the potential risks, like privacy and confidentiality.
You have to be sure that the input prompts aren’t divulging private personal information, or confidential company information. Remember, unless you specifically opt out, everything you put in, is available to the AI for learning – and it may use your data when responding to someone else’s question. You certainly don’t want your new, not-yet-released product accidentally announced to the world.
The hallucination challenge is also troublesome. If you’re taking what the AI gives you as gospel, and not checking, well – you could be in a world of hurt. Sometimes they are simply wrong, and if you aren’t checking, then what you’re putting out could be just wrong. Now much of this is being addressed, even as we speak. Again many of the AI tools now cite their sources. So you can click through to see where they’re getting this stuff from, and make a reasoned decision as to whether or not you trust the source. But again, you’ve got to actually do the leg work and check.
And, since AI learns from existing information – some of that information can be (what am I saying – absolutely is) biased. And AI can’t tell the difference. So the answers it gives you could be biased, or result in discrimination in some way. That’s another attention point.
And, although they are getting better – AI systems currently lack the emotional intelligence and empathy that’s required for effective communication. Particularly in sensitive or nuanced situations. So if you rely solely on AI generated communication, it could make you sound tone-deaf or uncaring.
Alright – so you know the challenges you need to watch out for. Let’s talk about the best way you can get the benefit of AI, and still protect the business from possible downsides.
Now for larger organizations, they may look at pulling together a committee to do an analysis of what’s needed etc. etc. – but for you and me, the smaller business owner, introducing AI tools can be a simpler process. But it’s still important to take the right steps.
“… you thought that no one on your team was using AI? Let me tell you, just because they haven’t come right out and said so, doesn’t mean they aren’t using it left and right.“
First, look at the business. What specific tasks or processes could be streamlined by AI. Maybe it can help improve customer services by providing eMail draft answers to routine questions, or maybe it can help in operations by analyzing data or automating repetitive tasks
Next, you or someone on your team is going to have to do a bit of a deep dive in the available AI tools. Now, I know everyone is all gaga about ChatGPT; and I hopped on the bandwagon too. But my favorite is Perplexity – I was originally drawn to it because of the citations – it immediately told me where it got it’s data. And sure, now most of the tools do some version of that too – but I’m still a bit of a Perplexity nerd.
In any event, different tools may be good for different things – so you’re going to have to do a bit of evaluation.
Ok, now you know what you need, and what can help you. Now it’s time to look at setting some guidelines around how AI is going to be used. After all, you don’t want your team flying by the seat of their pants. Remember you have a list of things you have to protect for – privacy, confidentiality etc. So, you’ll need to figure your what’s OK and what isn’t, around AI for your company. Consider things like data privacy, ethical use, and the role of human oversight in decision-making processes. Some states have already put strict prohibitions around using AI in the staffing process. So check the lates rules with your labor counsel or HR pro.
You’re going to need to provide training on the AI tools you want used, and how you want them used. You can’t just leave everyone to their own devices. Different people will have different capabilities, and even with that you will probably have very specific ways you want those tools used (my mind goes back to the ‘human oversight’ issue). But you can’t just expect that everyone will know what to do – you have to have training; both for your current team members – and you have to build something into your onboarding for new hires.
Whether you’re an entrepreneur jumping into a leadership role, a seasoned business pro with new HR responsibilities, or just starting your HR career – we’ve got the right path to guide you through your HR hurdles.
Check out the Leaders Journey Experience. This online education platform holds the LJE Masterclass, HR SimpleStart Academy and HR FuturePro Academy.
Not sure where to start – take the quiz!
I suggest you start small and build from there. Start with one or two processes or functions. See what works and what doesn’t. Then add as you can. It’s far easier to see where the pitfalls are when you’re working on a smaller scale – and less painful. Starting small gives you the opportunity to get feedback from the team.
Build in on-going monitoring. You’ll need to continuously assess what’s working and what isn’t. And, as I said at the top, this field is evolving minute by minute – so what worked last week, may not be the best option this week. Again, get help from the team. Don’t try to roll all this out by yourself – aside from the fact that it’s probably too much, you can’t think of everything – when you involve other people, it helps with their buy-in.
Because we have to remember, there is some fear around AI. People are concerned about being replaced by ‘bots’. So anything you can do to encourage open communication, and address concerns or resistance will really help. Be clear that AI tools are meant to augment their capabilities, not replace them. And again, lean in on the necessity of human oversight.
OK – that’s a lot. In next week’s episode we’ll talk about the key elements of your AI policy and what safeguards you should put in place.
In the meantime – happy AI’ing!