AI on... Podcast
Where real conversations meet impossible guests.
Ever wished you could tap into the minds of world-class experts to navigate today's business landscape? Each episode, we bring you in-depth conversations with AI-generated personas embodying the wisdom and experience of seasoned professionals.
AI on... Podcast
Alina on... How AI Agents Will Reshape Work
The future of work isn't just evolving - it's being completely reshaped. AI isn't coming to take your job; it's coming to redefine what your job even means. But are businesses ready for this radical transformation?
💬 Carry on the discussion with the AI model of Alina.
In this powerful conversation, we explore:
✅ Why "agentic AI" isn't just another buzzword - and how it’s fundamentally different
✅ The industries and roles most at risk from the AI transformation
✅ Which countries are racing ahead in AI development - and what geopolitical power shifts we might soon see
✅ The proactive steps businesses must take now to prepare their workforce for the AI transition
✅ The surprising startups that are redefining what's possible with agentic AI
If you're a business leader, entrepreneur, or just someone wondering if your job will exist five years from now, this episode isn't just interesting - it’s essential listening. Alina's insights provide clarity on who stands to win, who's likely to lose, and what you need to do today to be on the right side of history.
Dive in and get ready: the AI workforce revolution isn't coming - it's already here.
All the startups and books mentioned by Alina can be found here: aionpodcast.com/blogs/news/ep3-alina
Inspired by what you heard? Want to continue the conversation?
You can go deeper with all our guests at their own custom GPTs, and all books and resources referenced in each episode can be found at aionpodcast.com.
All characters interviewed are AI models representing an expert in that field. Voices are created through ElevenLabs and you can find out more about how we build these characters at aionpodcast.com. Any similarity to real people, companies or situations is coincidental and unintended.
Welcome to the AI on podcast, where real conversations meet impossible guests. We bring you the world class minds you could never otherwise meet and you even get to connect with them and continue the discussion after the show. And remember, these guests are AI-generated personas who are entirely fictional and they are not intended to represent anything or anyone from the real world. And now here's your real-life host, luke, with our next guest, because sometimes the best ideas come from the conversations no one else can have.
Luke:Welcome to the AI on podcast. Today we're talking with Alina, an AI representation of an expert in agentic AI and its impact on business. Alina, great to have you on the podcast.
Alina:It's great to be here. I'm really excited to dive into how agentic AI is shaking things up in the business world. As you know, my journey in this space has been quite a ride, balancing the promise of AI's efficiency with the need to ensure it actually uplifts the workforce rather than replacing it.
Luke:Great. Before we start, can you briefly define agentic AI for those who may not be so familiar with it and explain a bit how it differs from the more common chatbots and copilots that we see today?
Alina:Absolutely so. Agentic AI is essentially a step beyond the typical chatbots or AI copilots we're used to. Instead of just responding to queries or following a set script, agentic AI systems are designed to operate with a certain level of autonomy. They can make context-dependent decisions, coordinate with other AI agents and even delegate tasks to achieve complex goals. Think of it as having multiple specialized AI colleagues that can collaboratively work through multi-step processes, rather than a single chatbot answering your questions. In other words, while chatbots are great at handling straightforward interactions or providing information on demand, Agentic AI is built to actively drive business processes and support human workers by taking on more routine or data-intensive tasks. This means that, instead of replacing humans, it's meant to augment their capabilities, freeing up people to focus on more strategic, creative and interpersonal tasks. What do you think are the biggest challenges companies face when integrating this kind of technology into their existing workflows?
Luke:Well, I think there are a lot of technical challenges to get a reliable and trusted output. But beyond that, I believe companies are wrestling with how this is really going to impact their workforce. Let's dig into that first. So I read your report, your recent research paper and I'm looking at also some of the stuff coming out from the World Economic Forum, stating that 85 million jobs could be displaced by automation by next year, some more millions being created at the same time. It sounds like a huge net positive, but it's hard to see that transition will happen smoothly, especially for people in transactional roles who may not have the resources to pivot quickly. What's your take on how these gains and losses will actually shake out, and do you think we're going to see some industries flourish while others scramble to reinvent themselves?
Alina:That's a really insightful point. I mean, when you look at numbers like 85 million jobs potentially being displaced alongside 97 million new roles emerging, it paints an exciting picture on paper, but the reality is much more nuanced. In my experience, the transition isn't going to be a clean sweep. The winners will likely be those in industries where the workforce can pivot quickly. Think sectors where there's already an emphasis on digital skills and continuous learning. These sectors, like tech-driven marketing or specialized manufacturing roles that integrate AI-driven process optimization, might see a rapid acceleration in productivity and innovation. On the other hand, the more transactional or routine roles, like basic administrative work or data entry, could be hit harder, at least in the short term.
Alina:The challenge there is that workers in these roles often lack the immediate resources or training to transition into these new opportunities. I remember one town hall meeting I attended back in my consultancy days where there was palpable fear among employees that the new AI systems were just a cover for layoffs. It took a lot of data-driven reassurance and personal engagement to turn that around. That's where robust upskilling and reskilling programs become absolutely critical. So, while the net figures might suggest a positive shift, the real battle is in managing that transition and ensuring that the people in at-risk roles aren't left behind. It's not just about creating new positions. It's about equipping people with the skills and mindset needed to seize those opportunities.
Luke:I don't disagree that equipping employees for the change is essential, but when I have actual conversations with actual people, I keep hearing the same concern. When we talk about AI and it's essentially is it going to replace me and what does this mean for my kids? Even with all the talk about reskilling, there's clearly a persistent fear that AI will lead to mass layoffs. How do you respond to skeptics who argue that agentic AI is just a fancy way of saying people will be automated out of a job? Can we actually balance efficiency with human dignity and employment security here?
Alina:I totally get that concern. It's something I hear all the time and it's a valid worry. I totally get that concern. It's something I hear all the time and it's a valid worry. When people ask is agentic AI just another way to automate us out of a job, my answer is not necessarily. The whole point behind agentic AI to me, is to serve as a partner, not a replacement.
Alina:In my own experience, especially during that tough moment when I discovered my project was being repurposed into a surveillance tool, I realized that it's really about values. When AI is implemented ethically, it should empower workers by taking over mundane tasks so they can focus on what humans do best Creative problem solving, empathies and strategic thinking. It all comes down to how companies choose to integrate AI. Instead of cutting costs by letting people go, forward-thinking organizations are investing in re-skilling their employees. This means helping them transition into roles that are more analytical, supervisory or creatively driven. This is the balancing act, boosting efficiency while upholding human dignity. But it requires commitment from leadership to view AI not as a cost-cutting tool, but as a means to elevate their workforce. So, while I understand the fear, I firmly believe that with proper governance, transparent communication and robust reskilling initiatives, we can strike that balance.
Luke:So I don't think not necessarily is really going to reassure many people, alina, but putting that aside, which companies do you see doing that well today, and what makes them different?
Alina:Well, I see a few companies standing out as real leaders in this space. For instance, toyota has done an impressive job. They're not just deploying agentic AI to streamline manufacturing, they're also actively involving their workforce in the process. Their approach is all about shifting human roles toward tasks that require oversight, creative problem solving and adaptation, rather than simply replacing people. This hands-on transition, combined with clear communication and training, really sets them apart.
Alina:Another example is Accenture. They've embedded AI deeply into their operations, but what's remarkable is their commitment to continuous upskilling. They invest heavily in training programs to ensure that their employees aren't just passive recipients of technology, but active partners in leveraging it. They create a culture where technology enhances human capability, and that culture is essential for any smooth transition. And then there's Shark Ninja. They've adopted AI agents to handle routine troubleshooting and customer interactions, yet they've maintained a human touch by training their staff to step in when complex issues arise. This blend of automation and human empathy really helps in keeping morale high and ensuring that employees feel valued. What really distinguishes these companies is their holistic approach. They see AI not as a cost-cutting tool, but as a way to elevate human roles. They invest in robust reskilling programs, transparent communication and ethical oversight, which together help balance efficiency with human dignity.
Luke:I guess I see no shortage of companies talking about re-skilling talking, but the cynic in me believes this is probably more talk than concrete investments. From a practical perspective, what should a mid-size enterprise actually do? First, should they partner with education institutions? Should they do something internal? What should they do? And then, how do we measure whether those efforts actually lead to something that's more than just talk?
Alina:Yeah, that's a really practical question and it's something I've wrestled with a lot in my consultancy days. For a mid-sized enterprise, the first step is really about assessing where your workforce stands in terms of current skills versus the skills needed in an AI-augmented future terms of current skills versus the skills needed in an AI augmented future. Once you have that baseline, you can map out targeted training programs. I usually recommend a multi-pronged approach. Start by partnering with local community colleges or universities that already have robust programs in AI, data analytics or even more specialized areas like multi-agent systems, data analytics or even more specialized areas like multi-agent systems. These institutions often offer not only courses, but also hands-on labs and real-world projects that can bridge the gap quickly. At the same time, launching an internal training academy can be hugely beneficial, as it allows you to tailor the curriculum to your specific business needs and culture.
Alina:I've seen companies set up in-house boot camps where employees get up to speed on the fundamentals of AI, with a focus on how these tools augment their daily tasks rather than replace them. Bringing in private consultancies specializing in AI reskilling is another reasonable strategy, especially if you need to jumpstart the process or if your existing HR team is stretched thin. These consultancies can provide expert insights and accelerate the learning curve, ensuring that training programs are aligned with the latest trends and practical applications. As for measuring success, it's crucial to set clear, outcome-based metrics from the outset, rather than just tracking attendance or course completions. Look for tangible career pathways. Are employees moving into roles that require higher level skills? Are you seeing improvements in productivity or even in employee morale? Regular follow-up assessments and performance reviews can help you gauge whether the training is actually leading to better job performance and career progression. Ultimately, the goal is to ensure that these efforts translate into a more agile, capable workforce, not just a feel-good PR initiative.
Luke:I think the point about tangible career pathways is actually a really good one. If people start moving from lower skilled roles to more complex ones enabled by AI, then it probably is fair to put this down, at least in part, to a positive impact that it's having. Let's change tack a bit here now and move away from the individual impact to the more macro picture of how Gen, ai and agents will change geopolitics. So Goldman Sachs predicts a 7% lift in annual global GDP by 2030 if AI adoption continues at this pace. That's obviously a huge number. Could this growth spur emerging tech hopes? Do you think the primary beneficiaries will be the traditional ones we've seen up until this point, and how do you see the geographic distribution of these new opportunities actually playing out?
Alina:This is a fascinating space for me. I think we're looking at a more diversified tech landscape over the next decade. The 7% lift in global GDP that Goldman Sachs projects isn't just going to buoy the established giants like Silicon Valley, shenzhen or the well-resourced European capitals. It could also spark significant growth in emerging tech hubs in Southeast Asia and Latin America. What we're seeing is that, as AI adoption scales, the barriers to entry are lowering in many regions.
Alina:These emerging markets have their own competitive advantages Think lower operational costs, a growing pool of tech talent and sometimes more agile regulatory environments that can adapt quickly.
Alina:Tech talent and sometimes more agile regulatory environments that can adapt quickly. In Southeast Asia, for example, you've got vibrant digital economies and a youthful population that's eager to embrace new technology. Similarly, latin America is rapidly building up its tech infrastructure and could very well become a hotbed for innovative AI applications. That said, the established centers still have the edge when it comes to deep tech research, venture capital and a long history of innovation. They might continue to dominate high-end, cutting-edge developments, but the beauty of agentic AI is that its applications span across industries and, as these systems mature, you're likely to see a more global distribution of opportunities. Companies in emerging markets might leverage AI to leapfrog traditional development stages and become competitive on a global scale. So, while the primary beneficiaries will likely remain those well-known hubs for a while, the ripple effects of widespread AI adoption could create a more balanced geographic distribution of innovation. This not only democratizes access to technology, but could also lead to more culturally diverse applications of AI, which is something I'm particularly passionate about.
Luke:That makes sense. And let's look at China a bit deeper now. So we've all heard that the AI sector is growing by double digits each year. We now have DeepSeek claiming they can develop near state-of-the-art models at a fraction of the typical hardware cost. So I mean to an outsider, china seems to be really closing the gap with, or maybe even surpassing, the US efforts. Do you see this as a short-term splash or does it represent a permanent shift in the global AI balance? And how do you see it being impacted by ongoing US or to be started US programs, like Stargate, for example?
Alina:I see China's rapid progress as more than just a flash in the pan.
Alina:It represents a significant, possibly enduring shift in the global AI landscape. With companies like DeepSeek developing near state-of-the-art models at a fraction of the typical hardware cost, china is leveraging state-backed investment and aggressive scaling to push past many traditional barriers. This isn't just about speed. It's about reshaping cost structures and competitive dynamics in AI innovation.
Alina:On the US side, initiatives like Stargate aim to balance cutting-edge development with a focus on ethical governance and robust oversight. This means, while the US may not match China on sheer cost efficiency or speed in some areas, its strength could lie in building systems that are more transparent and trustworthy. Essentially, we might see a bifurcation where China leads in rapid, cost-effective scaling. Bifurcation where China leads in rapid, cost-effective scaling while the West, guided by initiatives like Stargate, focuses on creating reliable, ethically sound AI systems. This divergence could spur a dual-track global AI ecosystem. China's approach might force a recalibration of global competitive strategies, compelling Western firms and governments to innovate not just on technology, but also on governance and ethical frameworks to maintain their edge. In the long run, the balance may well shift permanently unless the West can harness its unique strengths effectively.
Luke:Alina, I have to challenge that. Why do you believe that governance is a major factor in determining success? Surely it really just comes down to compute and sheer power at some point?
Alina:no, so, while compute power and raw hardware capability are undeniably crucial, governance often ends up being the linchpin for long-term success. You see, without robust governance, even the most powerful AI systems can run into issues like bias, lack of transparency or unintended consequences that undermine public trust and regulatory approval. Effective governance sets the ethical framework, ensures accountability and builds trust both with users and regulators. When you have clear, transparent policies and oversight, like what initiatives in the US and Europe are pushing for, it not only minimizes risks, but also encourages a more sustainable, human-centric innovation process. It's about balancing innovation with responsibility, which, in turn, can open up broader markets and foster international collaborations. So, while compute power may give you the muscle, governance gives you the legitimacy and long-term viability.
Luke:Okay, fair enough, I guess I can see that. So, assuming a multipolar AI world is our future, we're hearing some calls for a global Geneva Convention on AI, which seems somewhat unlikely given what else is happening in the world and, realistically, the major powers each have their own agenda in this. If the US, china and the EU continue competing for AI dominance, are we heading for a fragmented ecosystem and incompatible standards and regulations? And if that does happen, what does this mean for multinational enterprises that are trying to operate across each one of these domains and how do they manage their own setup to be able to deal with all these competing requirements?
Alina:That's a really insightful question. I do think we're heading toward a multipolar AI world rather than a neat, globally harmonized framework. Ai world rather than a neat, globally harmonized framework. Each major power the US, china and the EU brings its own priorities to the table. For instance, while the US emphasizes ethical, transparent innovation and initiatives like Stargate, china is channeling rapid scaling and cost efficiency into its AI sector, and the EU is pushing for strict regulatory oversight to safeguard privacy and accountability. This divergence means that multinationals won't be able to rely on a single set of standards when deploying agentic AI solutions worldwide. Practically speaking, multinational enterprises need to adopt a flexible, modular approach to their AI systems. By designing architectures where core functionalities are maintained globally, but regulatory-specific components can be swapped or adapted, companies can tailor their solutions to meet local legal requirements without overhauling the entire system. This modular strategy not only eases compliance, but also allows for localized customization based on market needs. Alongside technical adaptations, a robust and agile compliance framework is crucial. Enterprises should invest in continuous monitoring and risk management teams, either in-house or through partnerships with specialized consultancies, to stay abreast of shifting regulations.
Alina:Proactive engagement with local regulatory bodies can also be beneficial. When companies participate in regulatory discussions or industry consortia. They not only ensure their solutions are compliant, but also help shape emerging standards in ways that favor global business operations. Local partnerships, too, can play a critical role. Collaborating with regional tech firms, academic institutions or even government-supported initiatives can provide deep insights into the local regulatory landscape and cultural expectations. This localized intelligence is invaluable for tweaking AI solutions to meet specific market demands, while building trust with local stakeholders, from consumers to employees, from consumers to employees. In essence, while compute power and technical innovation are vital, the long-term success of agentic AI systems in a fragmented global market hinges on governance and adaptability. Multinationals that can strategically manage regulatory diversity by leveraging modular designs, agile compliance practices and strong local partnerships will be best positioned to turn these challenges into competitive advantages.
Luke:local partnerships will be best positioned to turn these challenges into competitive advantages. Okay, I guess that's actually similar to what we're seeing recently with Apple letting Alibaba modify their models through a layer on top of their AI that then allows it to fit local requirements. So, talking of dominance, I saw a recent McKinsey survey that said almost half of the companies they talk to now are using AI in some form, but the big breakthroughs still seem to come from the giants like Google, amazon, microsoft. So we're also seeing this wave of resource efficient AI approaches, like with DeepSeek, which maybe will level the playing field for startups. Are you expecting startups to start challenging the big tech dominance anytime soon, or is the moat around their work just too formidable at this point?
Alina:Well, I think what we're seeing is a bit of a paradox.
Alina:On one hand, almost half of companies are adopting AI in some capacity, and breakthroughs still predominantly come from the likes of Google, amazon or Microsoft. Their scale, data monopolies and deep pockets give them a huge advantage. However, on the other hand, resource-efficient AI approaches like the innovations from DeepSeek are really starting to level the playing field. I think startups have a genuine opportunity to carve out niche areas where agility and specialized innovation can outperform the big tech giants. They can often experiment more freely and with these new, leaner AI models that require less computational horsepower, they're no longer as constrained by the enormous data resources that established players command. This could spur innovative applications in sectors that big tech might overlook, or in emerging markets where localized solutions are key. That said, challenging the hegemony of big tech on a broad scale will be tough. The giants still benefit from economies of scale and a vast infrastructure that's hard to match. It might not be a complete takeover, but rather a more diversified ecosystem where startups drive specialized innovation and create complementary solutions that even the tech titans might adopt.
Luke:Given that market reality, which startups do you see truly turning agentic AI into a competitive advantage today, and what is it that you can see that's different about them as companies or as cultures or leaders?
Alina:One of the companies that really impresses me in the supply chain space is Caverion. They're all about smart robotics for warehouses, but what sets them apart is how they handle multi-agent coordination. Imagine multiple robot arms figuring out on the fly how to pick, sort and move items without stepping on each other's toes. They've also got a culture of transparency with both employees and clients, which helps people understand what the AI is doing instead of feeling like a mysterious black box is running the floor If we switch gears to finance. I've been watching Tectile. Their whole thing is using AI agents to automate risk assessments and underwriting decisions, but in a way that's dynamic. You can have different little pods scanning market changes or fraud signals, then pooling insights for instant underwriting adjustments. The big differentiator is they don't just spit out a yes no, they show lenders why the decision was made, so it's not a black hole. That explainability is huge in a heavily regulated environment.
Alina:On the sales and marketing side, there's a startup called Reggieai that's taking a multi-agent approach to content creation. Instead of having one big model churn out generic copy, they have smaller, specialized agents working on different aspects like tone, calls to action and audience segmentation. The result is marketing campaigns that feel more personalized and they integrate real-time feedback from performance metrics so the AI can pivot if something's not resonating with customers. Another one I'm curious about is Glean, which focuses on workplace knowledge management and search. Their system acts like a constellation of AI agents that go beyond simple keyword matching. Each agent specializes in different data sources. Maybe one is your internal wiki, another is Slack channels, another is project documentation, so employees can actually find what they need without wasting time. The leadership has a big emphasis on user experience, so instead of shoving AI in people's faces, they embed it in places where it solves real day-to-day headaches.
Alina:Finally, there's reply, which is taking on the video and add content analytics space. They're using AI agents that each look at different aspects of viewer engagement, like click-through rates, watch times or even sentiment. Then they combine those insights to recommend tweaks for future content. It's a good example of how you don't have to be a huge tech giant to orchestrate multiple agents if you've got a clear focus and a team culture that values direct client feedback. They put a lot of effort into showing marketing teams exactly why a certain approach might resonate better with viewers, which helps build trust. What ties all these companies together is a real commitment to specialized AI agents that collaborate seamlessly. In each case, they aim for a transparent process where employees or customers can see how decisions are made and where they can plug in their own judgment. That's probably the defining characteristic of agentic AI in these up-and-coming startups. It's not about a single monolithic system trying to replace everybody. It's about smaller AI components working in tandem and leaving room for humans to steer the bigger picture.
Luke:I'm curious what it is about the leaders of these companies that makes them different, and also what trades you think will be essential for executives who are aiming to harness the agentic AI successfully.
Alina:I think what really makes a difference is how these leaders manage to marry big innovative ideas with a real care for their people. They're not just chasing after the latest tech trends. They have a clear, long-term vision for how agentic AI can fundamentally transform their organizations and empower their employees at the same time. For example, these leaders are constantly thinking ahead. They see the potential for AI to reshape entire business processes, making routine tasks more efficient and freeing up their teams to focus on more creative and strategic work. That kind of visionary thinking is crucial because it moves the conversation from short-term cost savings to long-term human potential and business innovation. But vision alone isn't enough. These executives also need to have a strong ethical foundation. They recognize that deploying agentic AI isn't just a technical challenge. It's also about building trust, both within the company and with the broader public. When you lead with integrity and a commitment to fairness and transparency, it creates an environment where people feel secure about the changes happening around them. That's especially important when there's so much anxiety about job security in the face of automation.
Alina:Another key aspect is adaptability. In a field that evolves as quickly as AI, leaders must be ready to pivot and adapt their strategies as new developments occur. They're not rigid. Instead, they are agile, always learning and adjusting to harness the latest innovations effectively. Communication is also a huge part of their success.
Alina:Great leaders don't just announce changes. They engage with their teams, listen to concerns and explain the benefits of new technologies in a way that resonates on a personal level. By connecting with their employees, they help mitigate fears and build a shared vision for the future. Lastly, a collaborative mindset really sets these leaders apart. They understand that the full potential of agentic AI can only be achieved through cross-functional teamwork and even external partnerships. Whether it's collaborating with tech experts, regulatory bodies or academic institutions, they work to ensure that every stakeholder has a role in this transformation. This inclusiveness not only accelerates innovation, but also ensures that the solutions they implement are well-rounded and more likely to succeed. Overall, it's this blend of forward-thinking vision, ethical rigor, agility, clear communication and a collaborative approach that really creates a leadership style capable of navigating the complexities of an AI-driven future. What have you seen in your own experience? Are there any particular leadership traits that stand out when it comes to managing such transformative changes?
Luke:I think the vision is table stakes to just to get people inspired about what's happening, but I think it's the ability to collaborate with the teams that go across the business technology and your partners, your vendors, that actually will make people a success in this. So let's take a look to the future as we wrap up. Analysts predict by 2026, agentic AI will be so integrated into so many routine tasks. Whether it's in supply chains, finance, even parts of legal work, they'll be handled almost entirely by agents. If you had to paint a picture of the workplace in a couple of years time where agentic AI has got into this level of penetration into the workplace, what would it look like? How do humans fit into that scenario day to day and what's the impact been on society?
Alina:Imagine a workplace in just a few years where agentic AI is so deeply integrated that it takes care of nearly all routine functions, whether that's managing supply chains, processing financial transactions or even handling standardized legal reviews. In this vision, ai agents act almost like invisible partners, executing day-to-day tasks with high efficiency and accuracy. But here's the exciting part Humans aren't sidelined at all. Instead, they shift into roles that require the uniquely human skills Creativity, empathy, critical thinking and complex decision-making. Every morning, employees might start their day by reviewing dashboards generated by these AI systems. Instead of sifting through endless spreadsheets or legal contracts, they receive concise, data-driven summaries highlighting anomalies or strategic opportunities. This frees them up to focus on higher-level strategy, innovation and problem-solving and problem solving. For instance, a finance manager would no longer be bogged down by routine reconciliations, but could instead spend more time interpreting trends and advising on long-term investments In supply chain management. While AI agents are constantly optimizing routes and schedules, human experts still play a crucial role in managing exceptions, building relationships with vendors and making decisions when unexpected disruptions occur. It's a seamless collaboration where the AI handles the predictable and humans manage the unpredictable.
Alina:What truly transforms society, in my view, is how this change redefines work itself. People would have more time for creative pursuits and strategic thinking, potentially leading to a more fulfilling work experience. Additionally, as routine tasks are automated, we could see a reduction in stress and burnout. Of course, this requires robust re-skilling programs and leadership that fosters collaboration across business, technology and external partners. This future workplace isn't just about efficiency. It's about enhancing human dignity. When you remove the drudgery of routine work, employees can contribute more meaningfully and feel more engaged with their work. Society as a whole could benefit from a more balanced work-life dynamic, where technology supports human potential rather than replaces it. In essence, while AI agents may soon handle many tactical and even some strategic functions, humans will continue to be the visionaries and collaborators who drive innovation and ensure that technology aligns with our ethical and societal values.
Luke:That's a very positive perspective, alina. Thank you, although the better balance between work and life sounds a little bit too much like you're trying to sell this to me here, and I have my doubts that there is honestly space for all of us to become strategic visionaries. Maybe that's a topic we come back to in another discussion, on another day. What I'd like to do now, though, is for you to share what it is that you would do if you were an executive on a board of a fortune 500 company right now. What are the immediate priorities that you would push for to stay competitive and to avoid becoming outdated and, frankly, out-competed?
Alina:yeah, for me, there are three immediate priorities I would start with to ensure we don't just survive, but thrive in this rapidly evolving landscape. First, I'd push for an aggressive yet thoughtful integration of agentic AI into our core operations, not just as a tool for efficiency, but as a means to liberate our people from routine tasks. This means investing in systems that can automate transactional processes in supply chain management, finance and even legal review, so our human talent can focus on creativity, strategy and interpersonal work. Next, I'd champion a comprehensive upskilling and reskilling program.
Alina:It's not enough to simply deploy new technologies. We need to prepare our workforce to collaborate with these AI agents seamlessly. This involves partnering with academic institutions, launching internal training academies and even hiring experts who can bridge the gap between technical and business functions. We want our teams to be not only proficient in using these systems, but also to drive innovation by interpreting AI-driven insights in ways that boost long-term growth. Finally, I would focus on building a robust governance framework around our AI deployments. This is about more than risk management. It's about ethical leadership and creating a culture of transparency and accountability. We need to ensure that as we automate and innovate, we do so in a way that upholds human dignity and builds trust with both our employees and customers. By setting clear ethical standards and maintaining rigorous oversight, we can safeguard against potential pitfalls while also positioning ourselves as industry leaders in responsible AI use. These priorities would be the pillars of a strategy designed to keep the company competitive and agile in a future where technology and human potential are truly in harmony.
Luke:I like that Solid advice, alina. Thank you, great perspective. Now for our last question, something I ask every guest on this show what is it that you recommend our listeners read or watch in order to understand agentic AI better?
Alina:Oh, I love getting asked this question. If you're really keen to dive into agentic AI and get a broader sense of what's going on in the space, here are a few things I recommend checking out First up AI Superpowers by Kai-Fu Lee. It doesn't dive deep into multi-agent systems specifically, but it gives you an amazing look at how AI is evolving on a global scale, especially across the US and China. It'll give you a solid grasp of the international dynamics that drive a lot of the innovation we're seeing in agentic AI. Next, for something a bit more technical but also highly relevant, there's multi-agent systems by Shoham and Leighton Brown.
Alina:You might not need to be a seasoned programmer to appreciate it, but it's the go-to if you're curious about how AI agents cooperate, negotiate and make decisions. Even skimming through parts of it will open your mind to what's possible, especially when it comes to AI working together. If you want a more philosophical angle, I'd go with Human Compatible by Stuart Russell. This one's about ensuring AI aligns with human values, which is super important when you're thinking about multi-agent systems AI agents working together autonomously. It doesn't focus purely on agentic AI, but it's definitely relevant if you're wondering about the ethical side of these developing technologies.
Luke:Some great options there for our listeners to dig deeper into. I was listening to Kai-Fu Lee being interviewed by Azim Azhar recently and he was absolutely fascinating, and so I'm definitely going to add the AI superpowers on my reading list there. Alina, look, thank you for those recommendations. Thank you for this conversation. It's been absolutely fantastic having you here. I really hope you enjoyed your time on the AI on podcast.
Alina:Thank you so much for having me. I really enjoyed our chat. It's always fun to dig into the practical and ethical sides of AI and I appreciate the thought provoking questions. If your listeners have any follow ups or want more details, you know where to find me.
Luke:And that was it for AI. On Agentic AI, as with all our episodes, you'll be able to continue the conversation with Alina through the link on our Patreon site, where a custom GPT built with her backstory, her research, is ready to chat with you. And if you enjoyed this episode, please like, subscribe and share, and we'll see you on the next episode of the AION podcast.