The Future of Work: HGS on Adaptability, AI, and Human Advantage

[00:00:00] - [Speaker 0]
Let's be honest. Security often sounds complicated. Firewalls, segmentation, access policies. It's a lot. But NordLayer takes away, and in less than ten minutes, you can set it up across your business and protect every employee connection.

[00:00:20] - [Speaker 0]
And they do that by blending VPN zero trust access and malware protection into one clean dashboard. Yep. No hardware, no long calls with support. Just toggle it on and get back to work. For me, it's the kind of practical solution I wish existed when I first started working remotely.

[00:00:41] - [Speaker 0]
So if you've been waiting for a security tool that your whole team can actually use without needing a degree in network engineering, NordLayer is it. So please head to nordlayer.com/ tech talks daily, and don't forget to use the code tech daily dash 28 for 28% off. Welcome to AI at Work, a podcast which is part of the Tech Talks Network. And in this podcast, we're gonna venture into the transformative influence of artificial intelligence inside the workplace. And our discussions will focus both the remarkable breakthroughs, but also the complex challenges of integrating AI into our everyday business functions and workflows.

[00:01:30] - [Speaker 0]
And my guest today is the CEO of HGS. He sits at the intersection of large scale operations, customer experience, and AI adoption. So today, we're gonna get specific. What roles are actually at risk? Is it data entry, admin, basic accounting, simple customer service roles?

[00:01:50] - [Speaker 0]
And which are the new ones that are beginning to emerge? Maybe prompt engineering or AI engineers, ethics specialists, hybrid ops roles that all blend process with product. And I wanna explore why fewer than one in 10 firms are meaningfully using AI, and yet 40% say they plan workforce reduction in five years. So how much of that shift is really about corporate strategy, profitability pressure, and competitive signaling. Well, if you lead teams, you'll hear practical ways today on redesigning jobs so that humans and systems complement rather than collide.

[00:02:30] - [Speaker 0]
So if you're early in your career, we'll map the skills with staying power, complex communication, ethical reasoning, creative problem solving, and the kind of judgment that no model out there can assume responsibility for. Because today, we're gonna have a grounded conversation about risk, about opportunity, and the choices leaders can make right now to safeguard access, mobility, and meaning at work. So let's dive in now. So a massive warm welcome back to the show. We have spoken in a previous life of people listening hearing about you for the first time.

[00:03:08] - [Speaker 0]
Could you just tell everyone listening a little about who you are and what you do?

[00:03:14] - [Speaker 1]
Neeling, glad to be, back on the show. My name is Anshuman, and I am the CEO of HGS, UK Limited. HGS is a company that looks at leveraging people and technology to provide better customer experiences. And, I live and work in London. So it's glad great to be back in the show.

[00:03:43] - [Speaker 0]
Well, thank you for sitting down with me again today. And when we're talking about AI and technology, there's so many different, opinions and concerns and other people excited and about the opportunities. But for people out there of any age that might be listening today, how concerned should they be when, about AI when they're entering the job market? From your vantage point, because you've you've seen a lot of here, where do you see this, heading, and and what should they be concerned about?

[00:04:17] - [Speaker 1]
So it's a very interesting question because, you know, we have been discussing about implication of AI at entry level jobs for a while. Now AI in general has been act in actual commercial use since at least for last ten years. If you remember, you know, Google playing chess or Netflix making recommendations or weather prediction. They were all AI use cases. But I think what caught people's imagination was what ChatGPT could do and the poems it could write.

[00:04:56] - [Speaker 1]
Now I think that create some degree of panic because, up until that point in time, the rationale was that AI can only be used for some very specific you know, statistical uses. And that completely changed with OpenAI and ChargeGPT. Now I think, in my opinion, the concern should be measured rather than panic because if you look at some of the history, and MIT had done some research on this, and it said that close to 60% of the jobs represent new type of work that has been created since 1940. And so that's almost, you know, more than majority of the jobs are new jobs that have come about in the last sixty years. Now this is not to say that, you know, we should ignore it because AI revolution presents some unique challenges.

[00:05:56] - [Speaker 1]
And the World Economic Forum created a report which said that, close to 40% of employers expect that their workforce will reduce, especially in areas where AI can automate tasks. And I think the key point here is there's an economist called David Autor who has said that it's really about timing. So if we look at the time between 1940s and eighties, the job displacement averaged about 17% while jobs were created at 19%. But in following years, the job displacement started to overshadow the job creation rate. So the question really today is about for people who are entering their career today, they should really focus on adaptability rather than obsolescence because what history has taught us is that your ability to leverage technology to do the job that exists today better or find new jobs that come along with it is where you will make the difference, and that's what we should be focused on.

[00:07:13] - [Speaker 0]
100% with you. And I think AI is often referred to as a copilot, something that works alongside teams to ensure work gets completed in ways that neither AI or humans could do as well on their own. The magic is when they work together. But getting your foot in the door, getting that early entry role, that seems to be one of the areas that's bigger, most affected by AI. So I think you hit the nail on the head there with adaptability and why that's so important.

[00:07:43] - [Speaker 0]
But from what you're seeing, what impact is AI already having on that that early career job market, those entry positions where people need to get their experience?

[00:07:54] - [Speaker 1]
So, it's interesting. When the initial reports started to come out, I was in two minds about it because I was thinking that, you know, some of them might actually be job displaced by technology. But, equally, we had just come out of pandemic, and there was a massive tech spending that had happened over pandemic. And then we were hit with economic slowdown, and that would always result in slowdown of jobs. But there was a interesting report that came out of ADP, which is one of the largest payroll processor, in the world.

[00:08:33] - [Speaker 1]
And what they realized is that the entry level jobs were actually slowing down. So while the job market as a whole was slow, it was the entry level jobs that were the hardest hit. And there was still some degree of growth in more in jobs that required you to have some degree of experience. So what that meant is that, yes, there were jobs in the market, and they were actually growing, but not if you are at an entry level jobs. That was concerning.

[00:09:04] - [Speaker 1]
And, I think JPMorgan, the bank, published, you know, a report which said that close to 6% of college graduates were not being able to, you know, you know sorry. There was an increase of 6% as far as the college graduates were concerned in their ability to find jobs. And the jobs that were hardest hits were the one that came with a degree of simplicity that did not require a lot of interpretation. For example, data entry jobs, admin and secretarial jobs, basic accounting roles, customer service representatives dealing with simple customer queries. So while this was happening on one end, equally there was evidence that you know, there are new entry level jobs that are coming in.

[00:09:59] - [Speaker 1]
You know, we have heard about prompt engineering roles, for example, that have grown almost 200% year on year in UK with average salaries of you know, close to 70 to £80,000. And so these, in a way, are displacing the jobs that probably paid half of it. And these are job areas that are growing very, very rapidly. I think some of the shift is also geographic and economic. So different parts of the world are growing differently in terms of job losses versus the gain in jobs.

[00:10:34] - [Speaker 1]
But I think all of this prompts back to the fact that, therefore, there is a need to look at, the government policy and what we are doing within the education system to make sure that people who are coming out of colleges or schools looking getting into the job markets are sufficiently trained to leverage the new tools that are coming out in the world.

[00:11:01] - [Speaker 0]
So many great points there. And I think for many people listening, if we do sit down and think about it, many people are doing jobs that just didn't exist twenty years ago at all. So the yes. Old jobs will disappear, but there are new jobs coming along. And you've mentioned earlier in our conversation that nontraditional roles, more and more of them are emerging.

[00:11:22] - [Speaker 0]
So are there any examples that stand out to you that that maybe are accessible to early career professionals to to give a bit of hope and and maybe shine a light on the opportunities that could be arising over the next few years?

[00:11:36] - [Speaker 1]
I think while there has been a lot of euphoria about AI technologies, specifically generative AI technology that can produce, prose or poetry or images. I think their accuracy levels are still, I think, 1020% of the time, these large language models get it wrong. So most businesses are still not comfortable using AI or generative AI technologies on complete autopilot. And that opens up a set of roles where you require human in the loop. But if I just break it down into, you know, three large categories of roles, there are the AI specialized roles or new roles like, you know, prompt engineers, as in how do you design the right way to ask a question to a to an OpenAI chat GPT so that you get the right kind of answers that you are looking for.

[00:12:43] - [Speaker 1]
You have jobs like, people who need to train AI or check the quality that, that it generates, or focus at the ethics of how some of these models are being developed. So these are extremely specialized roles, and therefore, their salaries are maybe twice of what a traditional graduate or entry level jobs do they command. Then there are roles that fall into the second bucket, which is the human AI collaboration, which is how do you leverage AI in context of a business or how do you use it to create more innovative ideas, whether that's within marketing or creative, know, disciplines or leverage generative AI to create content, whether it's for training, marketing purposes, and do it meaningfully so that you are not producing what is now being termed as work slop, which is the garbage that gets produced when you give half thought through or half baked prompts to AI. And last set of roles that are coming across are really about augmenting the traditional roles. So the business we are in, part of our business looks at you know, serving customers via contact centers.

[00:14:12] - [Speaker 1]
And you can serve a customer so much more better if you're you are leveraging the outputs that are coming from AI as to how the customer's last experience was or what else can be you know, you can look to cross sell or an intervention that you can make. Similarly, how you can leverage AI to optimize a process, etcetera. So I think those are the broad three buckets of roles that are coming in. Now I think they come with their own challenges and and advantages. So prompt engineering, for example, requires linguistic and creative skills as opposed to deep programming skills.

[00:15:00] - [Speaker 1]
And you would expect that somebody graduating from the college may not know the latest programming language, but they'll be well versed in linguistics or creative. AI trainers, for example, will need a better know how about the business than inability to write large language models and so on and so forth. So some of these new emerging roles are either require either require you to have good understanding of capabilities from the past that are still relevant, or they require you to have capabilities in the core of your business where you can leverage or use AI to just augment what you are doing. So I'm I'm I'm actually optimistic about the the new roles that are emerging, both which are completely new or ones that are human in the loop kind of roles.

[00:15:59] - [Speaker 0]
And from everything you're seeing and hearing out there, how much of the AI will replace entry level jobs narrative that we see in our news feeds? How much of that is about the technology itself versus corporate strategy and corporate priorities, etcetera?

[00:16:17] - [Speaker 1]
It's so this is this is a very interesting question and one that we debate very often.

[00:16:25] - [Speaker 0]
Yeah.

[00:16:25] - [Speaker 1]
If you look at a technologist pure play you know, purist's viewpoint, they will say, you know, well, the technology exists, and therefore, the scenario is very dystopian, that everything will get replaced by a technology like AI. But I think the reality sits somewhere in between because in a complex ecosystem, there are other forces at play, which could be political or corporate strategy or just, you know, socioeconomic. Now I think there was a survey that was carried out where there were less than 10% of firms who said that they were looking at AI to use it in the mainstream of their business in 2025, and the number of such firms were 10% or thereabouts. But in the same report, about 40% of the employers said that they intend to reduce their workforce because of AI in the next five years. Now you see that see the difference between the ambition or intent or what people are saying versus what they are doing.

[00:17:43] - [Speaker 1]
And if you if you go back in time and look at what happened during industrial revolution, there was technology that existed, but employment shifted from agricultural to more manufacturing, not because the machines immediately replaced all the farm workers, but because some of the companies said that, you know, the future lies in manufacturing and not in agriculture. And some of those companies moved to manufacturing. So I think in that sense, a mix of how different actors in a social context play out against each other within the realm of policy, regulation, market dynamics, and companies' own desire to move into more competitive or differentiated sectors will play out as to how much mainstream AI adoption impacts jobs. Some of the industries that are more pressed for profitability will see more AI getting adopted sooner. Investors and activist investors specifically will put pressure on companies to, you know, deliver more profitability, and you'll see more automation or AI adoption happening as a result of that.

[00:19:05] - [Speaker 1]
And I think lastly, we spoke about competitive positioning. Some companies will aggressively follow a strategy that is informed or powered by AI in order to develop a competitive position. And that's not new as in if you look at some of the tech companies like Google or Netflix, we use those products because they are great at giving you the right result or the right, you know, recommendation, and most of it is powered by So I think it'll be a think it'll be a mix of it. Sometimes what tends to happen is that the corporate narrative tends to oversimplify the narrative saying that, you know, the technology is ready. We can automate so much more, and we can become so profitable so fast.

[00:19:59] - [Speaker 1]
But I think it'll be a mix of the corporate strategy, socioeconomic realities, competitive pressure, activist investors, and all of this will be an outcome of that.

[00:20:12] - [Speaker 0]
And although AI does look set to automate many traditional entry level roles, I am a solutions, not problems kinda guy. So for business leaders listening, what long term impact might it have on leadership development and talent pipelines? And and what should organizations be doing right now to mitigate those risks and keep that pipeline flowing?

[00:20:34] - [Speaker 1]
Know, if I just go back early in my career, some of my early work was really designed around either doing research or creating alternate options or elaborating on a piece of work that somebody who was more senior had done. Now, almost all of this is now possible either through generative AI, generative design for a computer to do. And it'll therefore start impacting the entry level jobs very quickly. I think it'll it'll it is likely to do the following. That, you know, one is that there'll be a loss of apprenticeship opportunities.

[00:21:18] - [Speaker 1]
So entry level roles where people learn industry specific knowledge and the professional norms, that gets impacted. Entry level positions also often provide access to candidates without extensive networks or advanced degrees. The institutional knowledge transfer from the master to apprentice will take a hit. And people usually develop a sense of judgment having progressively handling more complex scenarios or tasks over a period of time. So I think these are the, I think, three, four areas that are likely to get impacted.

[00:22:03] - [Speaker 1]
And, you know, this is this is not new in many ways. If you, you know, just go back twenty, twenty five years when the computer revolution took place, low wage occupation jobs were lost while you had some of the higher paying ones that, gained traction. And over a period of time, whether these were universities or institutions that invested in training their employees to use computers, saw better long term outcomes than the ones that simply replaced workers. So I think in that sense, if you think about mitigation strategy, I think the first one is to create new development pathways, find ways to create hybrid apprenticeship opportunities that encourages young people to come in your workforce and provide them opportunities that help them master the AI tools. You know, do more rotational programs, do more project based learning.

[00:23:03] - [Speaker 1]
I think there is definitely a case of redesigning entry level roles and redefine what it entails. And I think one of the things that even we are doing as a company is to invest in internal mobility. So we are looking at people who are extremely good at running the customer experience management business into roles that are around developing the new AI protocols or new AI models in order to do either human in the loop sort of work or developing new models that will fuel AI led products. So I think we should we need to, look at doing that.

[00:23:44] - [Speaker 0]
You are so right in what you're saying there, and I remember hearing very similar stories back in the nineties with entry level jobs and the, arrival of computers and typewriters disappearing, etcetera. This time around, it's AI driven workplaces. Is there anything else business leaders should be doing to preserve that that human element of the the work? We've talked a lot around the importance of having the human in the loop. Keeping that human side of things is incredibly important.

[00:24:11] - [Speaker 0]
Anything else businesses should be doing to keep that?

[00:24:14] - [Speaker 1]
We all go back and look in the past and and look for references and and in the history to see what what we can do. So I think one great example of this is, if you think about maybe two decades ago, when e commerce came into picture, companies had two departments, one that sold into traditional brick and mortars and one that just did e commerce. And therefore, digital or digital marketing or digital commerce was a theme that sat in one corner of your organization. But today, especially after pandemic, ecommerce has become a mainstay of most businesses. It's completely integrated into what you do.

[00:25:01] - [Speaker 1]
So is digital marketing. Digital marketing is not a separate department. It is what your marketing department does. And, therefore, I think business leaders will need to start looking at every job description to see what does the human in the loop version of this job look like in the future? And they need to start looking at redesigning those pieces of work.

[00:25:26] - [Speaker 1]
Keeping in mind that they do not over swing the pendulum in the wrong direction, but find ways to make sure whatever the way the work is being designed, it is still has the right checkpoint for a human for human judgment and not something running on automation. Restructure roles around human strengths. We see many examples where most of the incoming tasks, whether it's your, you know, customer support help desk or IT support help desk or finance help desks, your level zero, level one task can very easily be automated. Your more complex scenarios or in certain cases handling vulnerable customers require true human empathy. So redesign roles around traditional human strength.

[00:26:19] - [Speaker 1]
But most importantly, look at developing a learning and development culture so that you are constantly reskilling and upskilling your workforces. And change your metrics from being a pure play metric that just looks at how much did you automate to figure out metrics that look at what human impact did I create as a business internally and externally. So how does my customer satisfaction and human touch interaction correlates vis a vis a completely automated workflow, for example? What are my employees saying about my culture and my upskilling and knowledge building skills? So we need to start looking at what impact we are creating internally and externally.

[00:27:14] - [Speaker 0]
And for everybody listening, I think we've we've talked about today the importance of being adaptable, and that is whether you are a junior employee listening, entering the workplace for the first time, or maybe even someone middle age who are equally concerned about becoming dispensable as AI engulfs the workplace. Anything or any advice you'd give to people listening of any age to ensure that they don't become dispensable and and continue to be adaptable? Any other tips around that?

[00:27:45] - [Speaker 1]
There are some distinctly human skills that are not going to be, you know, automated or go out of fashion anytime soon. I think the first one of them, we speak about it very often. So while it's very easy for a tool to act empathetically or demonstrate signs of emotional intelligence, but those are all programmed. I think to be able to really create meaningful connections that an AI cannot replicate easily, I think, is skill number one because that will still go a long way. Ethical reasoning and judgment, complex communication, creative problem solving, I think, are great skills that people should be investing in.

[00:28:38] - [Speaker 1]
And one example that I often cite is that if you go to any image generating tool, whether it is, you know, Sora from OpenAI or, Nano Banana from Google, if you look at the outputs of images that look fantastic, They're all generated by AI, but the images that look fantastic and great versus the ones that don't. And when you look at the prompt that somebody typed in order to generate the image, you will realize the that the a great looking image was generated by somebody who had great visualization skills. They were able to describe that image that they were seeing in their mind's eye before they fed it into the the AI tool. So what really distinguished a great image versus an mediocre image, both generated by the same AI engine, was the individual's ability to describe it in a manner that the other person couldn't. So this takes you back to your ability, linguistic, creative, innovative abilities, which are very human, that are not going to go away anytime soon.

[00:29:57] - [Speaker 0]
And I'm curious to end on another positive note here. What aspects of entry level work are best suited for humans? And, also, which which functions is AI unable to replicate? And and how do you see AI shaping the future of entry level work? I appreciate it's about three questions rolled into one there.

[00:30:17] - [Speaker 0]
But, what what will AI not be able to replicate, and how how will things evolve?

[00:30:22] - [Speaker 1]
So I will probably pick up your second part of the question on the current limitation, and that sort of informs what the future holds.

[00:30:31] - [Speaker 0]
Yeah.

[00:30:31] - [Speaker 1]
Now AI is great at identifying patterns but often struggles to understand the comprehension of why it is noticing the pattern it did. It does not understand often understand cultural context or historic, you know, nuances. And in many cases, sarcasm, irony, or implicit communication. So, you know, there's a difference. Most AI models are essentially statistical models that produce an output which says, what is my confidence level in saying x versus y?

[00:31:13] - [Speaker 1]
So it's really about pattern recognition versus understanding something. The second limitation it has is between narrow versus general incidents intelligence. So if you look at AI, it does great jobs in very specific domain. So it's very most of the new tools that have come out, they're very good at predicting the next word. That's how they write a paragraph or a poetry.

[00:31:41] - [Speaker 1]
But can they therefore write a scientific paper? The answer is no. So so they are not the artificial general intelligence does not exist still. They do a very good job in very narrow functions of of doing things. You know?

[00:32:00] - [Speaker 1]
There are previous examples of AI being able to predict fraud in a credit card transaction or being able to predict what the temperature tomorrow would be, etcetera. And lastly, there are legal angles and things like, you know, accountability and responsibility. Now if we put the three together, I think the advantages humans have over all of this is about building trust, communication, and relationship, solving problems given a particular context, being able to think creatively, thinking about original ideas and innovative solutions, and most importantly, ethical and moral reasoning on whatever is being produced by AI. I think these skills will continue to be valuable, and I think that's where we need to focus on.

[00:33:02] - [Speaker 0]
And I think that is a powerful moment to end on. And for anybody listening that would like to carry on this conversation or just read more about what HGS is doing around this as well and, dig a little bit deeper on some of the topics we explored today. Where should they go? Where would you like to point everyone?

[00:33:20] - [Speaker 1]
So I think there are few interesting pieces that are happening. You know, of course, you are you you should all look at our website, ags.com. And there is a fascinating podcast which talks both the positive and negative side of technology. It's called your undivided attention. That's a great place.

[00:33:45] - [Speaker 1]
MIT has been, producing a lot of interesting research on, shape of AI and its real implication and just, discerning the noise versus real data that is supporting as to what AI is doing. I think those are the three things that I would suggest.

[00:34:07] - [Speaker 0]
Well, I will add links to everything, including the podcast. I urge everyone listening to check that out. And we last spoke five years ago, and we cannot leave it another five years before we get you back on to complete your hat trick of appearances. So let's aim for early next year and see how things are continuously evolving. But just thank you for joining me again today.

[00:34:28] - [Speaker 0]
Really appreciate your time.

[00:34:30] - [Speaker 1]
Thank you, Lee. It's been a pleasure.

[00:34:32] - [Speaker 0]
I think that conversation today cut through a lot of noise. Three big takeaways for me to carry into any next planning meeting that you might have. First, the entry level squeeze is real, but so is the opportunity. Because if AI does succeed in automating the repetitive layers, leaders then owe their organizations new apprenticeships, hybrid programs, rotations, and assist then own pathways where juniors learn with AI rather than competing against it. And second for me is job design beats job deletion.

[00:35:13] - [Speaker 0]
Embed AI into roles that don't exile it to a lab, and let systems handle the repetitive and the retrieval and reserve humans for the relational, the ambiguous, the high stakes tasks. Then measure what really matters. Not just handling time, but sentiment recovery, ethical escalations that are resolved well, and lifetime value that is protected. And finally, future proving is a skill stack, not a job title. Being adaptable, curious, having empathy, structured thinking, visual storytelling, and the courage to ask better questions.

[00:35:52] - [Speaker 0]
These are the things that should travel with you throughout your career. A big thank you to today's guest for a thoughtful no drama tour of what's next. But until tomorrow, stay adaptable, keep learning, and design work where people matter more, not less. And on that note, I'm gonna walk off into the sunset. So I'll be back again tomorrow with another guest.

[00:36:15] - [Speaker 0]
Bye for now.