Transcript generated by Podium.page


NOTE: There were 3 speakers identified in this transcript.

0:00:04 - Helen
Hello and welcome to the HR edit, the podcast from South East employers, taking a look at those common and often tricky HR questions. Today I'm joined by Michelle and Sarah and we're going to be discussing the use of AI. So should we start with the definition of AI? 

0:00:19 - Sarah
Yeah, I can jump in on this one. So AI artificial intelligence is a broad term that refers to any type of computer software that engages in human like activities. So that includes things like learning, planning and problem solving, and probably, in the context of what we're talking about today, it'll be a lot of that sort of data analysis stuff. Now we've all got AI in our houses and we use it daily without even knowing. So a couple of examples and I'm stealing this from Michelle, I'll say that now, Alexa, we've got Siri, I've got doorbell camera, I've got smart cameras, I've got smart plugs that power my Christmas lights, and obviously, you know, one of the most common forms of AI is actually auto correct on your, on your phones. So AI is all about that kind of artificial intelligence, that kind of learns from those things that we're doing every day and our habits and everything else. 

0:01:09 - Michelle
Yeah, I mean, I use it to turn my lights on. I know people who their lawn gets mowed by AI, their house gets vacuumed by AI. Kids are using it to help them with their homework. My favourite, my daughter, about this time of year starts asking how many sleeps till Christmas and gets a lovely message back from the big man himself. So you know, ai has been around for a very long time and we have got accustomed to having it. You know, lender helping hand in our day to day life. Yeah, I think you know. With the launch of chat GPT last year, which made a massive kind of headlines, you know everybody's going to be out with the job. That's where, you know, we're starting to think more about how can AI help within the workplace and I use the word help deliberately, because we're not going to replace humans in the workplace. 

The way we work may change, but an example that I heard recently was accountants. It used to be not even that long ago. Bookkeeping was all done in ledger books. It was all done by hand on a bit of paper. End of the page. Everything got totalled up and transferred to the next page. Then we move to Excel spreadsheets and now we have accounting software that sits on our phones where you scan a receipt. As you leave the checkout it's, it notifies. You know what it looks at the receipt. It pulls out how much you've spent. You can then tag it and it puts it against a specific expense code or, you know, you can email it off to your, your finance team as an expense that needs to be reimbursed. You know we still need the accountants. We still need those people to do the end of year analysis and to check that everything is the right way and create those end of year account statements. But the way they work has changed. 

0:03:12 - Sarah
Absolutely. It's a supporting tool rather than a replacing of people. 

0:03:16 - Michelle
You know we're at that point now where AI can do, on some things, 80% of the work, but the human still needs to take ownership of that piece of work and do that last 20%. So, you know, in our context we have been playing around with the free version of ChatGPT. We've had it write a couple of policies for us just to test out its capabilities and it's level of accuracy. And you know we have looked at the output and it's great. It's generated it in a matter of seconds, which you know for us it would probably take us, you know, a couple of hours to think about and construct the sentences, but actually what we're getting is a draft policy that just needs to be tweaked and you know the language made appropriate for our organisation. 

We might want to put a little bit extra in, but the bulk of the work, 80% of the work, has been done and we just need to do that finishing touches bit. And that's where AI these days is a fantastic support tool. But it's not at, certainly right now and probably hopefully, in my lifetime or my working lifetime, should I say it won't get to the point where we don't need me in my role because I bring the, I bring the expertise to know whether what ChatGPT or any AI product has presented is accurate and worth the paper it's written on, so to speak. ChatGPT is an interesting one, isn't it? 

0:04:49 - Sarah
Because you know, it's kind of, I think, when it first came on the scene, I think, like me, like anybody else is pretty scared of the fact that it might take over doing stuff. Well, interestingly, you talk about policies. I ran a social media policy. Through it the other day. I wanted to see what it would produce and actually, though it produced a structure, it was very, very top level. So, in terms of producing a policy like that for a member organisation, for a local government organisation, it was very top level. 

So you I actually then had to spend another half hour actually going through and applying, like you do. You know, you know, you know, god forbid any of us actually rely on it and think, great, let's do my work for me. There is actually that kind of that human touch and, you know, remembering that it is a machine learning system, it really is just crowdsourcing that information from mainstream media to automation systems and GLY, market and other learning systems. But all as we go forward, I get a lot of people thinking about how is that possible system? It really is just crowdsourcing that information from what is already out there, which is human input. So, again, a great supporting tool and, like you say, could save you an hour, but actually you've still got to go through and put that level of expertise, on that level of nuance. You know that that machines just simply can't do. 

0:06:08 - Michelle
Yeah, absolutely, and it's one of those things that you know. 

Like you, I wanted it to write an employment law brief note for me the other day and I asked it to write a short piece on the updates to the Rehabilitation of Offenders Act amendments that have recently come in, and I'm sitting there going. 

Well, I know that those amendments are about reducing the amount of time before a conviction is considered to be spent, which, for brief purposes, if you have an unspent conviction and an employer or potential employer asks you to declare your unspent convictions, you have to then tell them what those convictions are and the employer then makes the decision as to whether they have any significant impact on your ability to carry out the job. So, by reducing the period of time in which the conviction is considered spent, it effectively means that you know, instead of having to declare it for 10 years, you only have to declare it for five. Now, what GPT throughout was the previous set of updates, which actually referred to extending the spent period timeframes, and so, instantly, as the human input, I'm like I can't use that because it's completely irrelevant, and that's where that sort of human sense check is absolutely critical. 

0:07:37 - Sarah
And I think that's the thing about AI is we talk about it as this big thing that you know potentially could put us out of jobs, but actually it's creating jobs Because a lot of AI is crowdsourcing information, is crowdsourcing data. Now, people who put that data in, let's be honest, are humans and it might develop to the point where you know it is crowdsourcing from other machines, I suppose. But another kind of AI system that we use at SE is Power BI and that you know that is very rich in AI features. It has the smart data discovery, the visualisations. They're all AI systems that are pulling from the data that again weave input or we've gathered or whatever. You know that spreadsheet I'm feeding into Power BI to visualise that data to make a nice dashboard. You know I've still done the work there. So you know it might be that our jobs change slightly and it might be that part of our jobs get really much, much quicker. So instead of pulling together a pie chart or you know just something to visualise some data for a client or a customer, I can actually stick it into Power BI and, brilliantly, it will bring up what it thinks I might want from the data that I want, whether it is what I want or whether it's not. It actually is saving me time because you know it's given me that kind of creativity. 

Another example as a graphic designer myself, I have used Photoshop. Well, photoshop now has an AI generator fill feature. So if there's a photo or something that I, you know it's a little bit too cropped, or I want to take a lamp post out, I can literally hit that AI fill now and it will edit, remove image elements, you know, and it does that all by analysing the pixels. You know it can smooth skin. But that encourages that kind of creativity and allows you to do stuff that you could always do. I mean, I could always take out a lamp post in a picture. It just took me longer. 

So you know what these systems are doing is they're learning, learning our habits, learning what we want to do. You know Photoshop realises that lots of people edit their photos. Well, that's what the system is for. So how can we make that easier and quicker? So there are, there are lots of different things that we are using, and it wasn't till we sat down to do this podcast 20 minutes ago that I was like God. What do I actually use for AI in terms of business, but it's all those processes like Power BI, like Photoshop, you know, and the other Adobe suites, that everything's getting quicker. 

You know I'm emailing every single day. I do God knows how many emails every single day and I'm auto correcting. You know that. That's learning my habits with Word. Word does the same and actually words now analysing everything as I put it in and telling me how good my writing is, which you know I find rather critical and unnecessary. But again, it's learning about. You know, learning what makes a good paragraph, learning what makes a good sentence. 

Sarah, you've missed out this. Are you sure you're not writing an American? No, I keep telling you, it's English default. It always goes back to American. So there's loads of different things and in terms of our website, for example, you know I go on there now to update content. You know gives me the option. Do you want to generate this contact through content, through AI? And I'm like no, not really, I need to write what I need to write about our services, but it's there and it's doing it. I sold some stuff on eBay the other day. I didn't have to write the description anymore. I could just press the AI description and it wrote it. All I added was from a smoke free home. You know it's that. It's that kind of somebody is a supporting tool, it's a help, I think, and I think that's what we're seeing more and more of, rather than being something sinister that's going to come and take all our jobs. 

0:11:01 - Michelle
Absolutely. And if you think we've been using AI on websites for you know, the last couple of years, because if you think, if you go on any retail website these days, you know you pop onto curries or I think even Smith's toys now have it they have a chatbot, yeah, and you can communicate with AI and say I'm looking for this. You know what do you advise for that and you know most customer service centres now use it. You know, if you, if you want to speak to your utilities company, the first thing they do is speak to our, you know, speak to our chatbot. They might be able to help. And then, if they kind of go, I don't know what I'm doing, they will then ask do you want to speak to an actual customer advisor? 

0:11:41 - Sarah
Yeah, it's that kind of triaging, that filtering isn't it. 

0:11:44 - Michelle
Absolutely it's. It's taking out that initial contact and you know people will complain about that because you know when you've got to complain, the first thing you want to do is speak to an actual human being. But if you're trying to find out what time the shop is open or what time the phone lines are open, or how do I make a claim on my car insurance or can I book an appointment to have my eye test, eyes tested For that kind of query, that would take a human being. You know I've got to answer the phone, find out who I'm talking, to log into the system to book the appointment in, take the details, Whereas actually a chatbot can do it really, really quickly and that frees up that person, that human being, to do the more constructive, productive work. 

0:12:36 - Sarah
Absolutely, absolutely. One of my favourite and also most hated is the Amazon one, and I had to use it the other day because a package didn't arrive for my mum as the nominated prime person in the family, and it's fantastic, it was absolutely fantastic, but it just kept sending me around in circles and I prefer the thing out the window because I then I couldn't. It wasn't a. I mean this is a very boring story and I do apologise to all the listeners, but it wasn't a package that was directly from Amazon, it was from a seller and you know it just kept sending me around in circles and but it's fantastic when it works. I mean, I'm just thinking about I was just thinking about our SEGWARP cybersecurity. 

That is absolutely fantastic with in terms of AI intelligence, because what we have now are these machine learning systems that are then machine learning defences. They can spot holes in network defences, they can recognise cyber attacks, and they can, you know, in terms of local government and you know, and private business as well. You know it's a really quick way to kind of find these issues and it alerts somebody I mean, again, it alerts a human being and we have to go in and then stop that, stop that attack. So, again, another kind of support tool, I suppose. But it's, you know, in terms of local government. 

Then I think what we're already seeing is those kind of, in broader terms, is that kind of traffic management systems. You know, they AI learn to sort of divert traffic and help analyse traffic flow and, you know, and helping the planners then and the traffic management people To then decide. You know, maybe we need to make this one way, maybe we need to do this. There's so many examples workforce planning, you know, all that kind of data stuff, all the OIS stuff that we do, the organisational intelligence service, all of that sort of rich data then is allowing people to make data driven decisions. You know, strengthening decision making within local government. I don't know if you I mean in terms of HR. We, I suspect recruiting is probably an area where you can scrape off LinkedIn, those particular talent. 

0:14:35 - Michelle
Yeah, I mean, I think there's a couple of things I want to kind of just kind of pick up on there. We're talking about machine learning and learning from you know the data that is out there, and that comes with its own risks. You know, by learning from what's out there, it's learning potential bias and there's there's sort of two stories that sort of linked to this and this is one of them was someone programmed an AI entity and put it on Twitter. The idea was that, you know, this AI entity would learn about human interaction from Twitter and generate its and kind of improve its own personality. It was done as a scientific experiment, so it was monitored and within a week they had to take the entity down, and the reason was it was analysing Twitter, and we know the content that gets put up on Twitter. It became, effectively, it became an evil entity. It picked up on all the negativity, all the the vitriolic conversation, and that became its personality. Hence it came down very, very quickly because it literally, within sort of two or three days, it had picked up that that was how humans interacted, which is a sad reflection on society these days, but we all know that that's what happens on a lot of social media platforms now because you're hidden behind a screen, you're not actually seeing the person eye to eye and people feel that they can say things that they wouldn't say to a person's face, and particularly on platforms where there's a level of anonymity with their handle names, they're completely hidden from society effectively. So there's sort of that inherent risk of learning the bad stuff from the data that's out there. 

The other aspect of it is and I use this example when I am talking to people about the negativities around AI if you look at the medical profession and it's getting better, it is improving. But the medical textbooks that our medical students still study from today are predominantly based on white male anatomy and white male reactions to viruses, to bacteria, to drugs, and we are beginning to recognise that that doesn't work because women have a different reaction to drugs to men. People who don't have white skin conditions manifest in a different way. You know prime example heart attacks. Women experience heart attacks in a different way to men. I can't remember which rash or which disease it is, but there's one where it manifests on white skin very clearly, but if you look at the same rash on darker skin it looks completely different. So the medical profession is having to adapt and do the research and update all the training to reflect the fact that you need to be aware. But at the moment, the majority of the information out there based on white male anatomy and physiology. And so if you put an AI presence into a medical scenario, whilst they are less likely to make a mistake and I use inverted commas on that because if you're going by the textbook they're not going to make a mistake but if that textbook is wrong to start with, it still has the impact on the individual that they're not getting the treatment they need. So, taking that into a local government HR workforce, there are lots of options for us using AI chatbots to deal with those. How do I book my annual leave? What was this year's pay offer? What's my leave entitlement? What's the maternity scheme? Those kind of policy-based questions that are just about fact-finding fantastic. 

The problem is when you're getting into AI making decisions, and we're talking a lot in HR at the moment around using AI for shortlisting in recruitment situations. We're talking about using AI to undertake job evaluation so that it takes the bias out of it, the human bias out of the job evaluation. We are talking a lot, actually about using AI for learning and OD. So those are probably the three key areas at the moment. There are lots of other options, as I say, policy generation, but you still need that human engagement. But if we take each of those in turn shortlisting in a recruitment situation Again the theory is taking out the bias and creating a sort of a non-judgmental analysis of an application, which is fine. 

But again it comes back to what has it learned from? Is there inherent bias in its decision-making process as to what is and isn't acceptable? So, for example, if you're asking for a specific qualification or equivalent experience, what does equivalent experience look like? It can distinguish whether you've got the qualification or not, but actually how does it demonstrate or how does it assess that you have the equivalent experience? If you don't have that formal qualification Someone with a disability it won't necessarily be able to pick up on that sort of adjustment that is needed. And, interestingly, those organisations who use the disability confident employer process and therefore guarantee an interview to anyone who meets the minimum requirements, who declares the disability, if you are saying, if you're programming the AI entity to say if they meet these 10 requirements and they've ticked this box to say they have a disability, then they automatically get an interview. But actually if you haven't programmed that in, you may be ending up with people being excluded from that shortest, unnecessarily or inappropriately, should I say. So it comes back to. Ai can do 80% of the work, but it needs that human check. So one option is anyone who ticks that they have a disability Once they've gone through the AI process. A human then goes back and says okay, let's just double check this. So it's not an automated decision. There is a human intervention at some point in that process. 

Similar kind of situation with job evaluation. Again, you can programme and it takes out the bias. You know it takes out. Oh, I know that, you know, I know this department and they're really struggling to recruit. Or you know, actually that department's run by my mate, so I'm gonna do him a solid and kind of push that up a little bit or score a bit higher. But again, it can't deal with that. It's that nuance. 

And equally, if you have a savvy manager who knows the job evaluation scheme, knows which words AI will be looking for, they will write that job description in a way that it improves the score and therefore hopefully gets that higher grade. Now, again a human looking at that job description will look at it and go. But I know what that job does, I know where it sits in the team and you, using that word feels as if it's egging up a little bit and will then go potentially and have a conversation with the manager and say are they really responsible, are they really accountable or are they just, you know, are they managing a process that supports that overarching delivery? So it's that sort of again I'm gonna say this a lot human intervention to do that sense check. That doesn't necessarily mean you can't use AI, but if you are going to do job evaluation using AI, it's worth just doing a sample sense check. So maybe every I don't know fifth or tenth job that goes through the AI job evaluation system gets pulled out and done by human and to see whether the scores are actually matching up, or have some kind of sense check at the end of the AI process that says this is where I would expect those jobs to come out. Have they come out? That one looks like a sore thumb that's come out higher than I would expect. So you can really start to sort of pull out those critical risk areas. 

The one thing to remember when you're talking about using AI for shortlisting in particular. But you know any person linked data, ai at the moment is an open source. So if you put data in, that becomes part of the wider knowledge that AI will use in the future to draw its sort of conclusions or to make its decisions. So, particularly if we're talking about using it around recruitment, there has to be a way that you can anonymise that data so that you're not putting someone's personal information into an open source platform and that's really, really critical, because what you don't want is to load someone's CV with their name and phone number and address into AI with their equal opportunities data where they went to school. You can imagine, given cyber security risks, that information gets put out into an open source platform. It is a mind field for scammers and criminals. A lot of people, a lot of security questions what was your first school, what was your maiden, what's your mother's maiden name or all of that kind of stuff. That sort of creates a security risk if you're putting that out and the individual who has applied for the job isn't going to necessarily be aware that that is what has happened with their data. The same will go for data that is of commercial sensitivity. If you're planning a restructure and you want AI to create the consultation documentation that you're going to send out to the trade unions and your staff, that in itself is sensitive enough. But actually, if you're putting in your business case and that has significant commercial sensitivity to it finances of the organisation or loss of a particular contract or things like that that goes out into an open source and can be pulled out by other organisations using the same AI platform, so some real potential risks associated with putting that sensitive data into AI. 

So the third area that I mentioned was around learning and organisational development, and this is one of the really exciting areas for me as a trainer, as an HR professional, as someone who likes to learn new stuff. We all have seen the training catalogs that training corporations produce. CPD is a prime example. Back in the day, you used to get an annual catalog and it was a nice brochure of all of their training courses. I need to sit there and you'd flick through and go that looks interesting. Or yeah, that look okay. Yep, I quite fancy doing that. And then you looked at the price and went maybe not. 

Ai has a brilliant function when it comes to mapping your career development plan, because all you need to do is go. Well, this is where I am now, this is where I want to get to. What courses, what qualifications, what learning would help me get from here to where I want to be, and that large corporations are starting to build this into their learning and OD platforms. To say, as an employee, tell us what your career aspirations are and AI will pick out your path to get to that point. And if you are currently in this career path but you want to find a way to change career, it can map that out for you and it can show you what experience you can gain, what knowledge you can gain that can help you move, make that step from I don't know finance to HR or HR to audit. It can map that across for you. And even if you know that your career is in HR, it can show you well, actually to be an HRD, you need to have this experience or this knowledge or this would be really helpful and this would be. It's not essential, but actually it would make you a well-rounded HRD and that's the sort of area that it's already in use in some organisations. 

But that is one of those things. I think for local government, we absolutely need to adopt that and join together with our neighbouring authorities and neighbouring organisations and share even cross-sector so sharing with the NHS our learning opportunities to be able to build careers and identify those pathways that enable us to developers individuals but actually improve our organisations. Because actually, if we're you know, if all of the local authorities and the NHS organisations and fire and rescue and police and all that you know, the public sector join together and create one big learning catalog, we don't all have to do the same thing. We can join with colleagues from different organisations, we can learn more from different organisations and we can develop our own internal specialist areas of development, but that doesn't impact on others who aren't in that special area. They can get it from a neighbouring authority. So it's just one of those things that you know. I think that has a real potential benefit to the public sector as a whole, but particularly for local authorities. 

0:29:37 - Sarah
Yeah, it's that career development you know, kind of personalised training and development program, I suppose. And sort of looking around before this podcast, that was one of the big areas that certainly in business that AI is going to become hugely helpful and, like you say that you know there's programs developing all the time and I suppose that kind of leads into the workforce planning stuff as well. There's a few workforce planning AI tools that are starting to sort of show up and they're great at analysing that workforce data, the industry trends, the growth plans, and can really help predict the roles which are going to need to be recruited to or, you know, depending on age of employees, all that kind of stuff. It takes all that into account so you can almost get that prediction for the year coming up, which is going to be a huge help in terms of workforce planning. 

0:30:25 - Michelle
Yeah, it really is In local government and beyond. 

0:30:27 - Sarah
Yeah, I know that there's a few examples of different ways that local government are using AI at the moment. There was a brilliant one at Surrey County Council where they teamed up with a specialist VR provider Virtual Reality and they kind of they created a digital model of Farnham Town Council and then participants, so local public could then fly around the town using jetpacks to kind of have a look and see what the different proposals bring. So there's a lot in the planning space as well. Same in Helsinki there's a somebody's built a virtual city of Helsinki and they can then start seeing, if they develop here, what it will look like, if they develop here, if they go higher than two stories or whatever. So there's massive potential in terms of AI, VR, AI kind of technologies. 

You know, certainly when you're looking at not just workforce but also that kind of environmental planning as well. It's a really exciting time, I think, to be in this kind of area at the moment, certainly as an employer's organisation, Because, you know, hopefully we can educate our local authorities in the southeast and work with our sister organisations as well, but to develop these kind of systems to help people understand. Actually it's not going to replace you, it's a supporting tool and how we could use it going forward, Certainly in local government, it's going to be huge, I think. 

0:31:51 - Michelle
Just picking up on that planning side of things when we know that there is a national shortage of planners. So actually how can AI augment that service? How can AI support those teams to take out a chunk of their work? Is there a way? And you know I'm not in that field, so I don't know the answer to this and I'm going to hold my hands up. You know I don't know but actually is there a way that AI can be used to do that? 

As you say, triage that first sift of actually that application doesn't meet the planning regulations, so automatically it gets flagged as a no. Human quickly double checks to make sure that it is a correct. But actually the main bulk of trawling through and you know, looking at the whole application is gone because actually AI says this doesn't meet this aspect of planning regulations. Ok, I just need to check that bit. Fine, no, you're right. Reject Again. It could get to the point where that doesn't need the human check on every application. 

It. You know, when you first introduce this process you check kind of every application that gets flagged as rejected. But actually after you have confidence, after you develop that confidence that the system does understand what it's doing, you reduce it to a sample. You know, one in every 10 applications gets viewed by a human being. Just to you know, do that sense, check and quality control what AI is doing. Well, that frees up a huge amount of time for those planners to go and look at the. You know the really challenging, complex, controversial applications and work. Do more work with developers and people who are submitting the applications to make sure that the applications are in line with regulations and, you know, meet the needs of the community or, you know, meet the needs of the area. So it's things like that. Again, it's taking away the. I don't like using the word mundane, you know, but I suppose it's taking away the everyday tasks so that that professional can focus their time and energy on the more strategic, the more useful and impactful work. 

0:34:17 - Sarah
And I think you know sort of taking a step before that, when you're thinking about planning as well, that kind of chatbot system that says do I even need planning permission for this conservatory? You know a lot of people don't know. I don't know, to be honest, and I've seen iterations of it. There's absolutely some local authorities that are using this, but unfortunately not all local authorities are using this and I don't know if this is. It's a case of being, you know, risk averse. You know not trusting, not trusting the software, but you know there are those sense checking that you can do, and you know there's a lot, of, a lot of councils that have apps where you can report stuff. We'll take it a step further have the app but also have the chatbot, you know, to help, to help that engagement, but also to free up resource, because we all know that there are huge strains on finances in local government. So you know there are ways of using AI as a first, as a sort of a gateway, as a triage system for your customer services, for your planning teams, for your, you know, benefits and housing teams, to kind of decide whether or not you do need to go through to that human interaction. 

And I know there's probably a lot of. You know, you think about the diverse population that each of our local authorities have got, and, of course, there are people that are always going to want to have that human interaction. But in the same way that we've got, you ring up and you press three to speak to somebody, or one to do whatever you know a chatbot it really is just that, but in a sort of keyboard, really, and it can be a hugely useful tool, not only for residents, but also for freeing up resources, and I think that's what it's all about. It's about that implementation and using AI in a way that's not going to cause more work, but it will actually refree up time and resources, because, at the end of the day, the biggest cost to a local authority is going to be the resource of its people. 

We don't have the people we can't. So I find it quite exciting, and certainly when we're talking about data driven decision making in local government as well, that kind of decision making that is backed up with a mountain of evidence, because it's there and because we can use it. But actually AI helps us get to that point. You know, if you make this decision, this happens if you make this decision, this will happen. So I think it's a massively useful tool and I think there is a yes, be mindful, you know, be aware of risks, but also let's let's with open arms. We should be in local government, you know accepting this technology and using it to its potential. 

0:36:42 - Michelle
Yeah, absolutely. I mean. I think the key is AI is not going to replace people, but it will replace people who don't embrace the use of AI. 

0:36:53 - Sarah
Yes, and what we don't want in, you know, as an employer organisation supporting local authorities in the southeast, we just don't want to be falling behind on that and I think, it's a really important part of our work, as well as that organisation, to to be up on all this stuff and to, you know, highlight those potential uses and the potential advantages and disadvantages of, you know, the uptake of some of this stuff. So a lot more work for us to do, I think, at SE to get to that point to be able to advise, but it is very exciting, absolutely. 

0:37:26 - Helen
Michelle, sarah, thank you so much for sharing those insights into AI. There is a lot to think about there. If you are a member of South East Employers and have any questions or queries related to the use of AI in your organisation, you can always contact us at advice@seemp.co.uk or for any other topics of HR advice. You can also visit the HR edit webpage at www.seemp.co.uk/theHRedit for any resources related to any of our episodes. This is our final episode of season two. Thank you so much for joining us and we hope our episodes have helped with those tricky HR questions. Be sure to check back at our previous episodes from this season and season one for more advice on those common queries. Subscribe to the podcast on Apple Podcasts, Spotify or wherever you get your podcast for notifications of when season three will be released in the new year, and visit the SEE website at www.seemp.co.uk for more information on how the South East Employers team can help support you with anything HR related. Thank you for listening and we'll see you in the new year. 

Transcribed by https://podium.page