As artificial intelligence and automation reshape the workforce, fear of displacement is driving widespread uncertainty. From strengthening employee trust to building human infrastructure and AI literacy, this conversation with Beena Ammanath, Global Head of the Deloitte AI Institute, offers business leaders insights into what it takes to responsibly implement AI and foster workforce confidence. In this episode, you’ll learn: • What trustworthy AI looks like in practice. • Why human-centered adoption and AI literacy are critical to success. • How forward-looking companies are equipping their workforces for AI-powered transformation.
As artificial intelligence and automation reshape the workforce, fear of displacement is driving widespread uncertainty. From strengthening employee trust to building human infrastructure and AI literacy, this conversation with Beena Ammanath, Global Head of the Deloitte AI Institute, offers business leaders insights into what it takes to responsibly implement AI and foster workforce confidence.
In this episode, you’ll learn:
Resources from this week’s episode -
https://www.shrm.org/content/dam/en/shrm/topics-tools/research/jobs-at-risk-brief.pdf
Subscribe to Tomorrowist to get the latest episodes, expert insights, and additional resources delivered straight to your inbox: https://shrm.co/voegyz
---
Explore SHRM’s all-new flagships. Content curated by experts. Created for you weekly. Each content journey features engaging podcasts, video, articles, and groundbreaking newsletters tailored to meet your unique needs in your organization and career. Learn More: https://shrm.co/coy63r
[00:00:00]
Jerry: I am Jerry Won. Welcome to Tomorrowist, where we explore the trend shaping the future of work. AI is evolving at lightning speed, and with it comes a looming question, will technology enhance the human workforce or replace it? New SHRM research reveals that 19.2 million US jobs face high automation displacement risk.
As headlines warn of mass displacement. How can leaders design AI strategies that prioritize people? Not just productivity. Today, we're joined by somebody at the forefront of answering that question. Bina. Amina is the global [00:01:00] head of Deloitte AI Institute and award-winning author and a global thought leader on ethical AI Bina.
Welcome to Tomorrowist.
Beena: Thank you so much for having me, Jerry. I'm looking forward to our conversation.
Jerry: You know, we can't go a day or maybe even half a day without hearing about ai. And there's obviously good things about how it's helping us, uh, become more product productive, creative and, and find new ways to, you know, advance our, our business and, and other agendas. But also there are other parts of it.
You know, we hear about, uh, even as, as recently as this weekend, a major technology company. Um. Announcing that due to AI or with the advancement of ai, that they're going to shift their labor focus, meaning they're gonna decrease, uh, their labor force or change where they hire. And so people are having mixed feelings about ai.
There are people that are obviously very excited about it and it's making our work and our lives a little bit easier. But, you know, naturally and understandably. Uh, folks and organizations who are worried about it. And [00:02:00] so, uh, in your view and your research, and you've written books on this, and you are the forefront leader on this, why has AI sparked such deep anxiety around worker displacement, particularly compared to past waves of other technology or technological, uh, automation.
Beena: Yeah, I think the easiest way to think about it, Jerry, is little look at back at the history, right? Um, the first industrial revolution was a lot about looking at, you know, making physical labor easier, where you didn't have to use. Physical, just pure physical strength from humans to get work done. That led to a whole different way of working, and there was jobs displaced, but it also created a lot of new jobs, but, and completely new products, right?
Like I don't think Jet engines would have been invented if we, you know. Pre-industrial era. So we have now looked, you know, with ai, we are looking more at that human intellectual layer or the, the brain part of our body humanness that is getting, uh, [00:03:00] you know, getting additional strengths, getting augmented.
And that's why there is fear because we are in this time of uncertainty. There's a lot of unknown, there's a lot of headlines and not everybody understands ai. So there is. That fear of jobs getting displaced, which is real and which is going to happen, but it's also gonna create new jobs. And I think the opportunity for our generation, Jerry, is to define those new opportunities proactively.
And I think you do a lot of that with your podcast right here.
Jerry: How do we define this balance between being human-centered and leveraging AI for what we know it can be, whether it's, you know, obviously we know, and then we've heard about the saving time and doing things faster and, and now with the evolving conversations around AI agents, technology, you know, uh, AI actually doing our jobs and, and thinking, um, and, and why is it important?
To, to make sure that we never [00:04:00] ignore the human impact of the AI and making it a human-centric conversation. And every time we talk about ai,
Beena: Let me start with a very simple example. You know, and this is something I actually faced in my career early on when, when I was build, looking at building, uh, AI models, machine learning models to look at, um, x-rays. And you come up with a come up with, uh, what, what the prediction would be based on the x-rays.
And the idea from a data scientist perspective was we are making their jobs faster, right? We are making it easier for them to do more X-rays. You're making them more productive. Right. And that I'm a technologist by training, so I was looking at. You know, solving it purely from a technology perspective, but from an x-ray machine operator's perspective, in their head they're thinking, okay, you know, AI is gonna free up 50% of my time, but it's not like humans are going to suddenly start breaking more bones.
And, you know, my queue is gonna get filled. What exactly am I supposed [00:05:00] to do in that 50% of freed up time, right? Uh, I'm certainly guilty of it, but many leaders, you know, brush, brush something like this under the fact that, you know, AI is going to make your job faster, easier, but don't go to that next level of.
What are, you know, what does this new change go role going to look like? What are they supposed to do with the freed up time? Or does everybody's job, you know, do we just go from eight HR days to four hour days and you know, your salary gets cut in half, that's just not gonna work, right? So you really need to be able to look at that next level of depth of the human impact, and that only works.
If you start with what's the problem you're going to solve for? What are the side effects of it, what's the impact to the workforce and how do you address it proactively? In this extra example, really, you know, in addition to building this, you know, a hundred person accurate algorithm, it would help to have the team also focus on what [00:06:00] is this human worker supposed to do in the freed up time?
Instead of just saying, you have freed up time, which is what a lot of leaders do. Have been doing, and I think that conversation needs to shift a little bit.
Jerry: Yeah. You know, there's conversations about time, but as you mentioned too, there's also conversations about sort of what do we do with the savings, right? And I'm using savings as a very broad term. You
know, if in this, you know, x-ray technician example, it costs half, um, half the amount of time, does that mean that the company pockets that difference?
Or is there some compensation going to the worker? How do you reinvest that to make sure that there's more technology? And, and so it's, it's a really interesting thing and I, I love the fact that you're reminding all of us to keep that at the forefront because while organizational leaders are tasked with thinking about and making decisions for what's good for the organization and the future, I.
Employees are naturally, as human beings as we are thinking about how does this impact me? [00:07:00] You know? And then there's, um, anecdotal stories that we hear from our friends and our neighbors and things that we see on the internet of, Hey, am I training this AI to replace me one day? Am I. Going to this conference to learn about something that's going to, uh, you know, help me not be here next year.
And so it's this delicate balance that I think people are excited about, the advancements that AI is bringing to our daily jobs, to eliminate some of the mundane, as you mentioned in the x-ray example, to make us more accurate without, you know, uh, eliminating the human bias in some of these things. Um, and so what are organizations doing right?
And, and where? And then what are some things that we can all learn from, uh, on things that they're not doing so well?
Beena: Yeah. And look, we are living in this transitionary time, right? You know, there's a lot of change happening and there's a lot of, you know, existing skillset that might get replaced or changed modified, but there's also new skill sets that needs to come to the forefront. You know, sticking to that x-ray example, right?
Um, you know, there's only so much that a data science or AI team [00:08:00] can do to. Or automate the process or speed up the process you need at the end of the day to get the next advanced level. You need deep domain expertise, right? You need the doctors, the medical background, the extra machine operators, domain knowledge, subject matter expertise.
So I think the easiest thing any organization can do. Sticking to this example is provide that base level AI fluency training, and that's really what I'm seeing as well. More many organizations now beginning to do is provide AI fluency training, AI literacy trainings to all of their employees. At the end of the day, AI is only successful in your organization.
If. If there is adoption and to drive adoption, you have to address the fear. And honestly, you know, the more you drive that AI literacy, you're going to get more. Ideas, more new product and solutions that come from the, the people who are, are doing their [00:09:00] jobs today. You know, they, once they understand what deep learning means or what is an AI agent, they're the ones who can come up, start, start thinking about the ideas of parts of their job that can be improved or done better or, you know, can lead to savings and can help define what the next.
Iteration of their role could look like, uh, there's, there's so much complexity to this topic, Jerry, but you know, I think if we just are mindful and thoughtful there, there is proactive steps that can be taken and many companies are starting with this AI literacy trainings.
Jerry: On, on that, and this is a great segue, so. For, for leaders who want to do it. Right, and again, this is such an evolving conversation. Maybe by the time our, our audience is listening or watching, listening to this or watching this, the technology's different or the times have changed. But what, what are some of the mainstay characteristics of a trustworthy AI system that can stand, that withstand the test of time or the evolving technology to make sure that we can [00:10:00] prioritize what's good for the worker and what's good for the company.
Beena: Yeah, you know, uh, the reality is the technology is evolving, so we are trying to keep up pace. With a change, rapidly changing technology, there's massive investments going into building the basics of ai, right? AI as a technology is still not mature. It's still very, very early days. Um, so I think when I think about trustworthy ai, I'm really thinking about it as the.
Side effects of the technology. Yes, you can drive more savings, you can get more ROI, you can generate new revenue opportunities, but it comes with side effects. Whether it is an impact to the workforce or new compliance, uh, violations or, uh, you know, the need for transparency, you know, how to address for bias in these algorithms.
You know, there are multiple side effects that come with using this technology in an enterprise environment. So I think when you. Think holistically about, uh, the technology and its [00:11:00] implementation. It's very important to set aside time to actually think through those side effects. Today we focus very much on, uh, you know, savings and ROI and, you know, uh.
It is impossible to think about all possible side effects, but at least you can get a good start and address it. 75% of the things that you might discover once a technology is rolled out. So, you know, I think it's more of a. Process change. Whatever project management tool you're using, or program management tool you're using, setting aside a percentage of time, and you have the smartest people in your company who are brainstorming on this new AI product idea.
Also think about what is this AI product once it's rolled out, once it's massively successful, what are the negative impacts? What are the side effects that it could lead to so that these really brilliant people who are building ai. Are also thinking about the ways that it could go wrong and address it as part of the development process, and [00:12:00] that's the missing step we give chase after the shiny object we chase after the cool new technology without taking a pause and taking that responsibility of also thinking about the impact that it could have when it's rolled out.
Jerry: So, you know, when we talk about sort of the impact on the business, you know, oftentimes when we think about ai, as you mentioned, the priority is on where the savings are, where the efficiencies are. Um, and so basically, you know, what are the tools and the data that, that can help us sort of progress things?
Um, what's being overlooked in that conversation about building the human infrastructure, needing to support meant that augment work, augmented work. Again, using the x-ray example, we're focused always on. How much time can we save? What can we have that person do in their free time? Or what are the savings that can be from maybe not having that person work full time, but the human infrastructure needed to support that?
Are there some additional things that have to be added to the equation in addition to the things that we are going to be saving?
Beena: Yeah, I think the human [00:13:00] infrastructure is absolutely crucial. I, I know we went through a frenzy a few years ago of, you know, uh, and you probably remember, uh, an article called Data Scientists is the Sexiest Job of the Century. Um, we went through a frenzy of, you know, getting hiring data scientists. But I think over time what, uh, we've realized is also this need for domain experts.
The need for under, you know, if you're building, um, say a medical product, having a doctor or a nurse or a trained medical professional at the table is super important. You are building a product for HR. It's not just going to be a bunch of data scientists. It ha you have to get in trained HR professionals as part of the product team.
So Jira with this need for domain experts and subject matter experts is a crucial part of the infrastructure as companies advance in their AI journey. Yes, at the beginning it's, you know, low, relatively low hanging fruit, and you can automate bits [00:14:00] and pieces. But to really take AI to the next level, to come up with the next idea, I think you need those domain experts at the, at the c at the table, to not only think about the product itself, but also the ways it could go wrong and the side effects it could have.
Um, so there's a lot around, you know, uh, we can talk about upskilling, reskilling, existing workforce. It all boils down to having that basic AI literacy to your existing workforce. The second thing I would point out, which we've not talked about is, you know, really thinking about. Proactively thinking about what are the new jobs for your industry?
We know there will be new jobs that created these AI systems are not just going to, you know, self maintain or, you know, keep self creating for your specific organization. So what are these new jobs? I think that effort needs to be put in into defining these new roles and not, you know, not just as a job description, but.
Thinking through what does the career pathway look like [00:15:00] for this specific role? Really going deep into thinking about what are the steps to get to that next level? What are the goals? What are success metrics? Um, I think that body of work is, needs to be done. Many companies are beginning to think about it, but it's still very early days.
Jerry: You know, AI is also a very subjective topic, meaning that it impacts different industries and different organizations in ways that are both unique to what that company does, but also across different pillars of, you know, industry and and size. We're seeing already the impact of ai, again, from a headline grabbing perspective.
Um, you know, over the weekend a large technology company announced that because of ai they're looking at downsizing their workforce. And again, depending on when you're listening to this conversation, that could be a number of different companies. Um, are there specific industries that are inherently better suited for human AI collaboration?
We're seeing it across the board right now. Uh, more so in the technology companies and [00:16:00] even. Uh, a few weeks ago, um, Spotify, the music company said, Hey, you need to use AI in your roles. Or, you know, like it's being infused into all of that. Um, will all industries be impacted at the same rate and at the same, you know, in, in the same ways?
Or are there specific industries that should be more focused on, uh, bringing in that human AI collaboration sooner because of the impact that the technology will have?
Beena: Yeah. And the, you know, I think there's, there's a lot to unpack here. One is AI as, uh, as a whole, right? We are seeing more traction in financial services and life science and healthcare because they have been traditionally data rich organizations, right? So the way that there is easily available accessible data, so it is much easier to build the algorithm so that there is, you know, that's where you're seeing a lot of advances, uh, in AI just because.
The foundation exists, but then there's also the other part to consider on the human aspect, which, uh, Jerry you mentioned is, [00:17:00] you know, I, I don't think there's going to be many roles that are not going to be impacted by AI in an enterprise. And, uh, I, I, I. This is my advice to anybody, right? It's not your, it's not, AI taking away your job.
It's going to be somebody who knows AI that will take away your job no matter whether you are, fresh out of college and starting a new job, or whether you're the CEO or board member, you are going to be using AI tools in your role role. So the person who uses. AI tools effectively for their current role is going to be, be the one that's promoted, is going to be the one that's recognized just because they're using the LA latest tools.
So in an organizational construct, it is absolutely crucial for everyone to not only understand the basics of ai, but proactively start learning about the tools that's relevant in your organization. What do I mean by that? Say if you're an accountant. Uh, working for A [00:18:00] CFO. Yeah. It is important for you to not only understand the basics of ai, but understand what is, what are some of the AI tools that's used in the finance domain for accountant?
What are other accountants using? What AI products exist for? AI accountants that the, for accountants that can be, that can help them do their jobs faster. So it's like, you know, when we started, uh. Excel, right? If you don't, if you are still using manual paper. And Penn compared to somebody using Excel and its formulas to do their jobs faster.
I think that's the transition that needs to happen. Uh, there is, uh, work to be done. There is no single playbook or handbook. A lot of it also comes down to the workforce to proactively learn and ask for their employers if they need that training. But at the end of the day, this transition period means you need to learn these new tools.
The reality is that AI is not going away. It's existed since [00:19:00] 1956, but we've, we are now seeing real impact. So whether you adopt that Excel or whether you continue with your paper and pen there, there is going to be a decision that you're going to make and that's going to impact your, uh, career in the future.
Uh, for, from an employer perspective, it is important to provide training on the AI tools. It's important to provide training on AI literacy, but more importantly, provide a sandbox. For employees to go and practice their AI skills to go and test out some of these new AI tools, uh, with a lot of AI tools being available publicly.
How do you provide a safe environment within the organization to, for employees to actually get their hands dirty and play with AI and get familiar with the tools that's
available?
Jerry: so we looked at it from an industry perspective. Um, what about from a size of organization perspective? You know, AI requires both a lot of investment knowledge, but also sort of the impact. If it [00:20:00] scales fast enough, there might be greater savings or greater efficiencies. You know, as we look at the vast variety in the size of organizations that are in the business world, um, are there certain, you know, is.
Should every, I think I know the answer to this, but, uh, uh, does every organization leader need to be thinking about it and at what point does it make sense to really think about it based on the savings? If you're, you know, leading a 10 person CPA shop that's gonna be a different AI impact than, you know, a large accounting company with hundreds of thousands of employees, um, how can organizations of a different sizes think about the impact of it and when should they be thinking about implementing some of the things, um, balancing both.
Uh, adoption, speed and benefits
Beena: I think today they need to be, uh, you know, no matter what size, what organization size you are, you have to be thinking about AI today. AI today is a co competitive advantage. If you, if you're not using it in your organization, no matter what the size you're going to be [00:21:00] left behind. It's just like for. You know, individual jobs.
It's a company that uses AI that's going to succeed, uh, more than the company that doesn't use AI at all. So it's, it's, it's, it really doesn't depend on the size. What comes down to the size and the industry is really how regulated is the industry. How, how much, uh, you know, how much of the data, uh, infrastructure do you have?
How much of the foundational infrastructure do you have? Set up that you can move at a, a faster speed towards ai. And that's where, you know, if you're a 10 person company and you, you may not have as much data, you might have to rely on an ecosystem to buy the AI tools of the shelf. Whereas if you're a larger organization and you wanna make it more fine tuned, you can actually train it on your own data.
So there, there are nuances to it, but I, I don't think any company today has the luxury to wait it out.
Jerry: I agree. And that's, you know, not, not to cause, you know, anxiety or, you know, panic amongst our audience, but it's something that you [00:22:00] should definitely be thinking about, reading about, playing with the different tools that are out there. And, and in almost every day there's a new tool, regardless of what function or industry you're in to sort of help with a, a, a different problem.
Um, we, we would love to get your perspective from your seat at Deloitte. Uh, what emerging practices are you seeing among. Forward looking companies choosing augmentation over simply automation.
Beena: A big thing is, you know, and we touched on some of this, right, having a a, a company-wide AI literacy training, making sure every employee in the company has access to the, to understand the basics of AI having a sandbox. Or a lab environment where you know, where they can go and play with what were the latest AI tool that they've read in the headlines, but in a protected, safe environment without any, any security leaks or concerns.
And then lastly, encouraging whether it is through having idea AI hackathons or idea hackathons. You know, really bringing employees [00:23:00] together to brainstorm, make them part of the AI journey, and create a structure and environment. Every employee can contribute to ideas of new ways of using AI within their role or for their organization.
That's where the best ideas will come from, right? It is. Once you put AI in the hands of your employees, they, and they understand that this is a tool and it can help me. That's when you'll hear the new ideas and really help the organization move forward. So I think we are seeing a little bit of that. Uh, the last one of, you know, really bringing the employee ideas together.
Uh, you know, whether it's through an AI day or AI brainstorming event, uh, that's, that's, um, that's another way to not only engage the employees, but get them on the AI journey.
Jerry: No, I I think it's great. Um, you know, 'cause this entire AI conversation that we're having today, and then, you know, the, the big sort of fear-based ones is about being left behind. You know, there are, [00:24:00] uh, individuals who again, think that AI is going to replace my job. There are entire industries who've done a lot of the data analysis work and their entire companies who do work.
In this regard for other companies. And so fear, fear is natural, right? As business progresses, as humanity progresses, and then we have to evolve. And we've talked about a little bit here already, but as we wrap this conversation, bina, what, what are some of the things that leaders can do and, and think about, um, you know, reshaping workplace culture.
Um, you know, to, to make sure that it, they've actually foster a sense of inclusion, learning, adaptability. You just talked about bringing others along, you know, through some of the, you know, hackathons and sort of things that we can do. But from a culture perspective, what can leaders really prioritize and continue to reiterate to make sure that they're not, that, that they're also gonna be a part of this next wave of journey.
Beena: Yeah. And here is an interesting, you know, observation that I've had, right? The technology can move at its own speed, right? AI tools are being launched every [00:25:00] single day, and there's new headlines every single day, like you mentioned, Jerry, but the adoption of the technology within the organization moves at the pace of the culture of the organization.
It is not at all the same. Speed. Right? So it's crucial for leaders to help drive adoption of the technology through some of the things that we've talked about, uh, but also, uh, getting them on their journey means having these conversations. Having the, uh, you know, encouraging employees to listen to podcasts like these or, you know, providing a stipend for buying AI related books and reading it and pro, you know, having an AI book club.
I mean, there's so many different ways that you can start getting AI into the vernacular, into the culture. And the more an the leaders can themselves show how they're using AI in their jobs, I think the more. They adopt, you know, employees will be more open to trying it. The fear is [00:26:00] absolutely real. And if you don't understand anything about what's going on with ai, it's very hard to accept that technology.
And I've personally experienced it where, you know, you can build the most accurate algorithm and nobody uses it. At the end of the day, it's a failed project. Right? So it is. Crucial to invest into driving adoption. You know, there's so many different ways you can do it and, you know, having an AI day, a simple thing, but bringing, bringing all your employees together for a virtual conversation on AI and make it, you know, make it open so that they can ask questions and you know, at top, top of mind questions, what were the headline might be.
Have that. Discussion pro, providing those platforms and forums to engage with leaders, um, is, is crucial. And it is okay in this scenario for leaders to also share, you know, where they are in their AI journey, how they're using it, and you know, also share that they're not comfortable as well if they're not, and how they, you know.
Creating a blog post on, you know, how they are, [00:27:00] you know, using AI in their daily lives. I think there's so many different ways, Jerry, that leaders can step up and drive this cultural change. Um, and I, I think most employees are also keen to get on this journey, uh, you know, to allay those fears.
Jerry: We, we talk about leaders' roles and organizational roles in not only shaping this conversation, but also, you know, um, what, what role do they have and, you know, the ethics and the accountability of it. Who should own these conversations? You know, and, and we have discussed that length. The fact that AI will touch every.
Aspect of a company regardless of function and then things. And so, you know, these conversations that are having, uh, that are happening inside organizations about ai. Who owns this? Is it the CTO? Is it a new role, the Chief AI officer? Is it the CEO? Um, what is your, what's happening now and where do you see this sort of evolution of ownership at the executive level going on the topic of [00:28:00] ai.
Beena: There, there is. Um, it depends on where you are on your AI journey. I think. Um, there, there are, and there are different models. There's no fixed. Playbook. Uh, I speak to a lot of company boards and, you know, even board members are getting on, on the, on the AI journey, right? Whether it's using AI for their board meetings, but also engaging with the CO to understand what the organization itself is doing.
So a few models, and you touched on it, right? Driving it as a in strategic initiative from the CEO side, um, having the CTO or the CIO expand their roles to look at AI or bringing in a brand new transformation kind of role. Chief AI officer, right? Like we had Chief Digital officers, when companies wanted to pivot to more digital, I think Chief AI Officer is a new role that comes in more as a transformation.
And this role could. Set, either reporting directly to the CEO or to the CTO or sometimes even the chief operating officer. So there are [00:29:00] different models here. There is no one size fits all. I think depending on, you know, how much, how far ahead you are in your AI journey, you might want to bring in a new role who's focused on, you know, just.
AI and looking at it holistically across the organization. Or you might want to start with small, uh, small, uh, projects, use cases within the organization and bring it u under the CIO or CTO. So there are different models to it.
Jerry: You've given us so much knowledge and direction on what organizations and and leaders can do, um, and. As with ai, there might be a little bit of what, what do I do next? What is the first thing that I can do? And so, uh, as a, as AI evolves, and, we continue to have these conversations, Beena what is the one thing that leaders should prioritize today to build more human centered organizations tomorrow?
Beena: Proactively think about humans, it's easy to think about the savings and the value impact. It [00:30:00] is absolutely crucial though to think about the human impact and place humans at the center of it. proactively thinking about that can drive some of the behavior changes that we talked about today.
Jerry: Awesome Bina. It's, it's been such an amazing conversation. I've learned a lot and I know our audience as well, and you know, the, the conversations like this. Will continue to happen. The topics will change, the technology might change, but as you have reiterated to our audience through this conversation, always make sure that you prioritize the humans in your organization, in your ecosystem, and make sure that that is always a priority part of the conversation around ai.
Uh, Bina, thank you so much for joining us on Tomorrowist, and we'll see you tomorrow.
Beena: Thank you for having me.