Leading IT in the AI Era: How to Build a Frontier Firm

Why 2025 Marks a Tipping Point for AI in Business

We are witnessing a once-in-a-generation inflection point in technology. In 2025, AI is no longer a future promise — it’s the engine of digital transformation. Business leaders are rethinking operations, workflows, and culture around intelligent systems. Microsoft calls this the rise of the Frontier Firm — where AI is woven into every process, and IT leadership must evolve from builders to orchestrators of human–AI collaboration.

82% of business leaders say this is a pivotal year to rethink strategy and operations, and 81% expect AI “agents” to be integrated into their company’s strategy within 12–18 months.

Adoption is accelerating fast:

24% of leaders report their companies have already deployed AI company-wide
Only 12% remain in mere pilot projects

In other words, the era of widespread AI in business isn’t theoretical or distant – it’s happening right now. This surge is backed by unprecedented investment and innovation. Stanford’s 2025 AI Index underscores AI’s breakout into the mainstream economy:

Total corporate investment in AI hit $252 billion in 2024
Private AI investment jumped by nearly 45% year-over-year

AI is no longer at the fringes; it has become “a central driver of business value.” I have spent over many years leading IT transformations – from early cloud migrations to large-scale cybersecurity initiatives – and I’ve never seen a technology evolve and be adopted at this pace. It reminds me of the advent of the internet era, only compressed into a much shorter time frame. In fact, some analysts compare today’s AI-driven shift to the Industrial Revolution or the rise of the internet in its magnitude. The message is clear: we’ve reached a transformational moment where every IT and business leader must decide how to adapt to AI’s new reality.

In this post, I’ll share my perspective – as an IT executive who has led digital transformation, cloud adoption, AI integration, and security programs – on what this “Frontier Firm” era means for IT leaders. We’ll explore:

The Frontier Firm concept defined by Microsoft’s research
Insights from Stanford’s AI Index and other studies
How to navigate the key shifts ahead

This is a personal take grounded in my own lessons (with company names left out) and aimed at helping fellow business and IT leaders lead with vision and courage through the changes to come.

What Is a Frontier Firm and Why IT Leaders Must Care

So, what is a Frontier Firm? Think of it as the prototype of the future company – one powered by AI at its core. Microsoft’s Work Trend Index defines a Frontier Firm as:

"A company powered by intelligence on tap, human-agent teams, and a new role for everyone: agent boss."

In simpler terms, these organizations seamlessly blend human talent with AI “agents” (autonomous software assistants) in every aspect of their operations.

3 Pillars of a Frontier Firm

1. Intelligence on Tap

Frontier Firms can buy and leverage AI capabilities on-demand, much like electricity from a grid. Advanced AI models and services are available as a utility – instantly scalable via cloud platforms and APIs. For IT, this means our role shifts from building every solution from scratch to integrating external AI services and managing “intelligence” as a new kind of resource. We ensure the business has access to the latest AI tools (whether from cloud providers or AI startups) and weave them into our systems securely. In my experience, having a strong cloud foundation is critical here. Our early move to the cloud in my previous organization turned out to be a key enabler for tapping into new AI services quickly when they emerged.

2. Human–Agent Teams

In Frontier Firms, AI “agents” join teams as digital colleagues, working alongside people. Rather than replace jobs, these agents take on tasks at human direction and augment what teams can do. Microsoft describes a future in three phases:

First: AI as an assistant
Next: AI as a teammate
Finally: AI autonomously executing entire workflows with humans overseeing

We’re already seeing phase 1 and 2 in action:

Customer support reps use AI copilots to draft responses
Marketing teams have AI assistants researching data

As these agents mature, org charts may be “upended.” Imagine teams where a mix of humans and AI report to a manager, or where a single employee can supervise a fleet of AI agents doing work. For IT leaders, this raises important questions:

How do we architect systems for human–AI collaboration?
How do we provision accounts and access for AI agents?
How do we monitor performance when part of your “workforce” is software?

It also means rethinking processes: many workflows will be redesigned around what AI can do best, with hand-offs between humans and AI. In one transformation project, I worked with operations managers to map out which steps an AI could handle (like generating a draft analysis) and which required human judgment. We treated the AI like a new team member during process design – a mindset that will become the norm in Frontier Firms.

3. Every Employee Becomes an Agent Boss

Perhaps the most profound shift is cultural. Every worker, from the boardroom to the frontline, will need to effectively delegate to and direct AI agents – essentially acting as a “boss” to digital assistants. Instead of just doing tasks, employees will increasingly guide AI to do tasks. This changes the skills required across the organization. As a leader, I see my role expanding to ensure everyone is trained to use AI tools and to foster a comfort with supervising AI outputs. According to Microsoft’s research:

28% of managers are considering hiring AI-specific team managers to lead hybrid human/AI teams
32% plan to hire AI specialists to design and optimize these agents in the next year or two

New roles are emerging (like AI workforce manager or prompt engineer), and every employee’s job is being redefined to include working with AI. For IT, enabling this means:

Providing user-friendly AI platforms
Offering training resources
Creating governance so employees at all levels can experiment with AI safely

In one of my previous organizations, we ran an internal program to train staff on using a generative AI tool for data analysis. The goal was not just to introduce a new tool, but to nurture an “agent boss” mindset where our analysts learned to delegate data crunching to AI while they focused on interpreting results. The experience was eye-opening: some of our employees who were initially non-technical became power users of these AI assistants when given the chance. It showed the vast untapped potential in our workforce once AI was made accessible.

In essence, the Frontier Firm concept signals a future where AI is woven into every fiber of the enterprise. For IT leaders, it’s a call to action:

Provide the infrastructure, platforms, and policies to support AI at scale
Supply “intelligence on tap” securely
Redesign workflows for human–agent teams
Empower every employee to leverage AI

The payoff is huge:

Frontier Firms, by blending machine intelligence with human ingenuity, “scale rapidly, operate with agility, and generate value faster.”

They can:

Handle surging workloads by filling the “capacity gap” between business demand and human ability with digital labor
Respond to market changes more quickly, because decision-making and execution are accelerated by AI

As an IT executive, this gets me excited. It’s our opportunity to elevate IT from a support function to a central role in rewiring the business for intelligence. But it also comes with challenges that we need to navigate carefully – which brings us to the key shifts we must lead our organizations through.

How IT Leaders Can Navigate the Shift to AI

Adopting AI at this depth isn’t as simple as flipping a switch. It requires navigating several major shifts in how our businesses operate. Let’s break down some of the critical shifts of the Frontier Firm era — and how we, as IT and business leaders, can lead through them:

1. From Building IT to Buying “Intelligence on Tap”

In the past, IT was often about building or customizing software. Now, with AI capabilities available on-demand, the mindset shifts to assembling and integrating intelligence. Off-the-shelf AI services (from computer vision APIs to large language models) can be rented as needed. This shift means leaders must become savvy in technology sourcing and integration. We need to evaluate third-party AI offerings, negotiate contracts and partnerships, and ensure our architecture can plug into multiple AI services. In my experience leading cloud migrations, having a flexible, cloud-based architecture made it much easier to experiment with new AI services – we could spin up an environment, connect an AI API, and test value in weeks instead of months. Leadership Tip: Encourage your IT architects and procurement teams to treat AI services as a new supply chain. Establish a strategy for when to build vs. buy AI solutions. For example, you might buy a proven AI service for translation or image recognition, but build in-house expertise for domain-specific AI that gives competitive advantage. Also, keep a close eye on cost and performance; “intelligence on tap” should be optimized like any utility – ensure you have monitoring in place for usage and ROI.

2. Human–Agent Collaboration and Organizational Redesign

Introducing AI colleagues will impact organizational structure and processes. Leaders should prepare to redesign workflows and roles to maximize human–AI synergy. One key concept I focus on is augmentation: AI agents are there to amplify human productivity, not just to automate. To lead through this, we must actively involve our teams in reimagining their work. For example, when we implemented an AI-driven ERP assistant, we invited our finance and operations staff to co-design how it would be used in their daily work. This collaborative approach helped employees see the AI as a helpful partner rather than a threat. We identified mundane steps the AI could handle (data entry, initial analysis) and freed up employees for higher-value activities (complex decision-making, innovation). As a leader, set the tone that AI is a team member. Encourage managers to pilot human–agent teams on small projects — perhaps have a “digital colleague” assist a project team and document what works and what doesn’t. Microsoft predicts that in phase 2 of AI adoption, these human–agent teams will become standard, and they even foresee new metrics like a “human–agent ratio” to balance oversight with efficiency. Be prepared to measure and adjust how work is allocated between people and AIs. Also, consider org structure: if agents take over certain tasks, do you need the same layers of management or can teams be leaner? It’s conceivable that the classic org chart will evolve into what Microsoft calls a “Work Chart” – structured around jobs to be done, not just departments. Leading through this means being willing to challenge old job definitions and empower cross-functional teams that leverage AI at their core.

3. Culture Shock to Culture Shift – Making Everyone an “Agent Boss”

Not everyone on your team will immediately be comfortable working with AI, and that’s okay. One of the biggest leadership challenges is guiding people through the culture change. We need to create a culture where using AI is as normal as using a spreadsheet. This involves education, trust-building, and policy-setting. In the Frontier Firm era, every employee must eventually “think like the CEO of an agent-powered startup” – taking initiative to direct AI tools to solve problems. To move toward that, I’ve learned the value of demystifying AI for the workforce. At one company, we held hands-on workshops for different departments, showing very practical ways AI could help in their specific jobs (for example, how a salesperson could use an AI assistant to draft a proposal, or how HR could use it to screen resumes). We also created an internal community for sharing AI success stories, which helped late adopters see peers benefiting from the tech. Leadership Tip: Invest in widespread training and make it continuous. Don’t assume a one-time session is enough; provide on-demand learning (videos, internal wikis, AI sandbox tools) so employees can build their skills over time. Equally important is addressing the trust factor. Many employees naturally have concerns about AI’s reliability or even fear it might displace them. Being transparent about why you’re adopting AI and how you intend it to augment (not replace) people is crucial. I made it a point to involve our Ethics & Compliance team to develop clear guidelines on how employees should use AI — for instance, when to keep a human in the loop, data privacy rules, and quality checks on AI output. Clear governance helps people trust that there’s a safety net. Microsoft’s research found that leaders are already far more familiar and comfortable with AI than the average employee, and they caution that bridging this gap:

“Will take more than access; it will require training, oversight, and a new way of working.”

In short, champion a learning culture where experimenting with AI is encouraged, and provide the oversight to use it responsibly. When employees see leadership actively using AI in their own work and sharing lessons (yes, I regularly show my team how I use a GPT-based assistant to summarize reports or draft plans), it sends a powerful message that this is the direction forward. Over time, your organization builds the muscle to be an agent-driven enterprise.

4. Balancing Innovation with Security and Ethics

Amid the excitement of AI, leaders cannot afford to drop the ball on cybersecurity, data privacy, and ethical considerations. If your company is to be a Frontier Firm, it will be heavily reliant on data and AI algorithms — which also means greater exposure to risks (from data breaches to AI biases). One lesson I carry from leading ISO 27001 security certification is that security enables innovation when done right. By establishing strong data governance and risk management early, you actually create a foundation of trust that lets you move faster with new tech. So how do we lead through this shift? First, embed security and ethics into your AI initiatives from day one. For example, when we started using a cloud-based generative AI service, our IT security team worked closely with the vendor to understand encryption and data handling, and we set up role-based access so only approved data could be fed into the AI. We also educated employees on what types of information were okay to input (avoiding sensitive customer data, for instance). This proactive stance prevented problems and reassured our compliance department – it paved the way for broader AI use because we weren’t reacting to incidents, we were preventing them. Second, keep up with evolving best practices. AI governance is a nascent field; frameworks like AI ethics principles, bias audits, and model documentation are becoming important. I made it a point that our data science team establish an ethics review for new AI use-cases, even if it was a simple checklist to start with. Additionally, consider certifications or standards as tools in your arsenal. Our achievement of ISO 27001 (information security management) turned out to be a big advantage when integrating AI. It meant we already had controls for access management, incident response, and supplier risk — all of which we applied to our AI vendors and projects. The board and our business unit leaders were more confident adopting AI knowing that those same security guardrails were in place. Leadership Tip: Partner your AI architects with your security and compliance teams. Have them design solutions together so that innovation and protection go hand in hand. This might slightly slow down the initial rollout, but it dramatically reduces the chance of a costly setback later.

Nothing derails digital transformation faster than a security breach or a compliance violation.

Remember, a Frontier Firm must earn trust from customers and regulators to fully leverage AI. By leading with a mindset that innovation and security are both non-negotiable, you ensure your organization can push the envelope sustainably. As Sundar Pichai, Google’s CEO, said:

“AI is more profound than fire or electricity” — but we must handle it with responsibility.

Lead by setting that example.

5. Navigating Uncertain ROI – Focus on Value and Iteration

It’s worth acknowledging that despite the hype, extracting real business value from AI is not automatic. Many executives are bullish on AI investment – 92% of executives in a recent McKinsey survey plan to increase spending on AI in the next three years – yet there’s healthy skepticism about return on investment. Stanford’s AI Index notes that AI’s impact on ROI is still being debated and not well understood. From my own journey, I’ve learned that you often have to prove the value in stages. When we first deployed AI in a critical business process (an AI scheduling optimizer in a logistics function), the initial results were underwhelming. We realized we had to fine-tune the data inputs and retrain people on interpreting the AI’s suggestions. Over a few iterations, the performance improved and eventually delivered significant efficiency gains — but if you looked at the first quarter, the ROI wasn’t obvious. This taught me to set realistic expectations: pilot projects might not show instant payback, and that’s okay if you’re learning and improving. Practical step: Set up metrics to track not just end results (like revenue lift or cost reduction) but also leading indicators of progress (e.g. adoption rates of the AI tool, reduction in cycle time for a process, user feedback scores). Celebrate early wins to build momentum, but also be candid about setbacks and lessons. By creating a culture that values learning over perfection, you encourage teams to keep pushing the envelope with AI. And always tie projects back to business outcomes — don’t do AI for AI’s sake. For instance, we aligned our AI initiatives with strategic goals (customer satisfaction, operational efficiency, etc.), which helped focus our efforts and gave meaning to the experimentation. This way, even if the exact ROI is hard to calculate in advance, you ensure you’re working on what matters. One interesting data point: only about 19% of companies in that same survey reported a significant revenue boost (>5%) from AI so far. The majority saw modest gains or none yet. This isn’t to discourage, but to underline that value from AI comes from diligent execution and iteration. As a leader, be prepared to steer your organization through trials and adjustments. Maintain support from top management by sharing both the promising data points and the improvements plan when results lag. If you stay committed, the breakthroughs will come – and those who figure out how to reliably translate AI into business value will be the Frontier Firms that leave competitors in the dust.

Before moving on, it’s worth noting one overarching leadership principle in all these shifts: people come first. No matter how advanced the AI, it’s the people – employees, customers, partners – who determine whether it succeeds. I remind myself of this daily. As one AI expert insightfully put it:

“The technology is leaping forward, but the people and processes take time to change.”

We must lead with empathy, clarity, and support for our teams as they adapt. When you do that, you create an organization that’s not only technologically capable but also resilient and ready to seize the opportunities of the Frontier Firm era.

Leadership Lessons from the Frontlines of Digital Transformation

Having led multiple digital transformation initiatives, I’d like to share a few personal lessons that might help others on this journey. These are moments and insights from the trenches that shaped how I approach leading IT into this new era:

Lesson 1: Technology is the Easy Part – Bringing People Along is Harder (and More Important)

During my career, I was involved in a major software rollout. We had the best technology deployed on time and on budget, but adoption lagged terribly. That experience ingrained in me that success lies in change management, not just tech deployment. Fast forward to recent AI projects, and this lesson rings true more than ever. In one case, we introduced an AI analytics tool for our sales teams. Technically, it worked great, but usage was low after launch. We discovered that many team members didn't trust the AI's insights and felt their own expertise might be undermined. It took concerted effort – workshops, one-on-one coaching, and showing quick wins – to turn that around. I was honestly surprised by the depth of emotional resistance, even among digitally savvy staff. It reinforced that as leaders, we must communicate a vision that ties the technology to individual purpose. I started framing the AI tool not as some fancy new gadget, but as a virtual assistant that could take over menial data crunching and free up time for salespeople to do what they do best – build relationships with clients. We also enlisted a few enthusiastic early adopters to mentor their peers. Gradually, trust built up and adoption rose, and soon the tool became a valued part of the team's workflow. The lesson here is humility: no matter how transformative a technology is, never assume people will automatically jump on board. Empathize with their perspective, involve them in the process, and lead with why it matters to them. This "people first" approach has become a cornerstone of my leadership style. Every time I plan a new AI or system implementation, I allocate as much effort to training, communication, and listening to feedback as I do to the technical work. As the Frontier Firm era accelerates change, this human-centric leadership will be even more crucial.

Lesson 2: Pair Bold Innovation with Rigorous Governance

One of the most important lessons I've learned leading IT in highly regulated industries is this: the more innovative your work, the more important your guardrails. Take our ISO 27001 journey as an example. We were pushing hard to modernize our digital infrastructure — moving critical workloads to the cloud, introducing AI-powered services, and accelerating time-to-market for healthcare apps. It was exciting. Fast. Mission-critical. But operating in a sector where trust is everything — especially in digital health — meant that speed couldn't come at the cost of security or compliance. So, we didn't treat ISO 27001 as an afterthought or a bureaucratic checkbox. Instead, we made it the backbone of how we innovated. We embedded the principles of risk management, access control, incident response, and supplier vetting into every new tech initiative from the start. It paid off. By the time we were piloting AI-based data workflows, we already had the frameworks in place to assess third-party risks, manage internal access, and prove compliance — even under tough international regulations. Our adoption went faster, not slower, because people trusted the foundation it was built on. The lesson? Governance doesn't have to be a brake on innovation. Done right, it's a launchpad. It gives your teams the confidence and clarity to experiment — without guessing what's safe, or what might cause problems later. When innovation moves fast, trust must move faster. And that’s something we, as IT leaders, are uniquely positioned to deliver.

Lesson 3: Embrace Lifelong Learning – for Yourself and Your Team

The rapid evolution of AI has humbled me more than once. Just when I felt I’d wrapped my head around deploying machine learning models, along comes generative AI and whole new paradigms of prompt engineering and agent orchestration. One thing I’ve learned is to never get too comfortable with what you know. As a leader, it’s important to stay curious and keep learning. I made it a habit to set aside time each week to read up on emerging tech trends (often from sources like Stanford’s AI Index or industry whitepapers) and even to play with new AI tools firsthand. This not only sharpens my own understanding but sets an example for my team. I remember one instance where I took an online course on AI ethics alongside some of my team members – it was a great equalizer and a team bonding experience, showing that learning has no rank. The surprise benefit was that it created a culture where continuous improvement was valued. Our employees started sharing articles, hosting lunch-and-learn sessions, and bringing fresh ideas to the table because they saw that initiative was welcomed. In my blog posts and internal talks, I often say: in the Frontier Firm era, the most important skill is the ability to learn new skills. Encourage your people to carve out time for learning, whether it’s formal training or self-driven exploration. Provide them resources – maybe give everyone a subscription to an AI learning platform or sponsor certifications. Also, be open about what you don’t know. There were times I admitted to my team that I was still figuring out how to best apply a new AI tool – and together we’d brainstorm and experiment. Such honesty can empower others to voice their questions and ideas. The days of the know-it-all executive are over; in a fast-changing landscape, we’re all students. Embracing that mindset has been one of the most liberating lessons of my career.

Lesson 4: Collaborate Beyond IT – Make Transformation a Team Sport

One more lesson I value is that IT cannot drive transformation alone; it truly takes a village. Some of the most successful projects I’ve led were those where we had a tight partnership between IT and other functions – a coalition of the willing, so to speak. For example, during a company-wide digital transformation in a consulting client, we formed a steering group that included not just IT folks, but also leaders from operations, finance, HR, and business units. This cross-functional team was golden. When we started rolling out AI-driven process automation:

Our HR partner helped devise a re-skilling program for roles that would change
Our finance lead helped quantify the benefits and secure funding
Our operations managers identified priority pain points to target with AI

This broad involvement meant that the transformation wasn’t seen as an “IT project” – it was a business initiative with shared ownership. One instance that pleasantly surprised me was when a non-technical operations manager became one of the biggest AI champions, because she saw directly how it could solve a logistical scheduling nightmare she’d had for years. She co-led the implementation with us, and her team’s productivity soared. The takeaway for me was to break down silos. In the Frontier Firm era, technology permeates every aspect of the business, so our approach to leadership must be inclusive. Bring different perspectives to the table early – you’ll catch issues that you in IT might overlook, and you’ll gain allies who will help drive adoption. It’s also important for messaging: when employees see a united front of leaders from various departments advocating for AI-driven improvements, they understand it’s a strategic priority, not just the flavor-of-the-month from IT. In practical terms:

Set up governance that includes other stakeholders
Create an AI council or task force with representatives from legal (for policy), HR (for training and workforce impact), core business units (for use-case identification), and IT (for tech enablement)

This collaborative leadership approach has consistently been a recipe for success in my experience, and it’s something I’d recommend to anyone looking to turn cutting-edge tech into real organizational change.

Each of these lessons was learned through experience — and sometimes failure. They shape how I lead today. I share them in hopes that they spark reflection in your own journey. Leading IT into the frontier isn’t just about mastering AI algorithms or cloud configurations – it’s equally about mastering communication, strategy, empathy, and learning.

A Step-by-Step Playbook for IT Leaders Embracing AI

Bringing together the concepts and lessons discussed, here is some practical advice I would offer to IT and business leaders aiming to navigate the Frontier Firm era:

• Craft a Vision and Roadmap for AI in Your Organization

Don’t adopt AI in a piecemeal fashion. Take the time to define a clear vision for how AI and automation will create value in your business over the next 3–5 years. Consult your business strategy – where can AI drive the biggest impact (be it customer experience, efficiency, new product offerings)? Then develop a roadmap that prioritizes a few high-impact use cases to start, and lays out the capabilities (data, infrastructure, skills) you need to build. I’ve seen companies succeed by treating AI initiatives as a portfolio managed with the same rigor as any strategic investments. Also, make sure to get buy-in for this vision at the highest levels – board and C-suite support ensures you have the mandate and resources to execute. Remember, as many as 92% of executives are ramping up AI investment, so if you can articulate a strong plan, you’re likely to find support. Your roadmap will evolve as technology does (expect to update it at least yearly), but having that blueprint is crucial to avoid chasing every shiny object and instead focus on what drives your competitive advantage.

• Invest in Skills and Create an “AI-Ready” Culture

The workforce won’t become AI-fluent on its own. Plan and budget for training programs to raise the AI literacy of your employees. This can range from basic AI orientation sessions for all staff to advanced machine learning courses for technical teams. Also identify power users or “AI champions” in each department – passionate early adopters who can help others. Encourage peer learning and knowledge sharing, perhaps via internal communities or hackathons. Culturally, celebrate experimentation and curiosity. One idea is to set up an internal innovation lab or sandbox where employees can play with AI tools on non-mission-critical projects. I’ve done this via a “demo day” format where teams showcase small AI experiments – it’s amazing how this energized people and surfaced new ideas. Additionally, update job descriptions and career paths to include AI competencies. Show your people that learning to work with AI is not just nice-to-have, but a core part of growth in the company. And crucially, address the mindset shift: communicate that AI is there to empower them, not to make them obsolete. By providing both the knowledge and the reassurance, you create a workforce that’s not afraid of AI but eager to harness it. As Microsoft’s research highlighted, bridging the gap between leaders and employees in AI readiness requires deliberate effort in training and new ways of working – make that a priority.

• Strengthen Data, Technology, and Security Foundations

AI prowess is built on strong fundamentals. Make sure your data house is in order – that means having the infrastructure to collect, store, and process large volumes of quality data. Consider modern data platforms (data lakes, real-time streaming, etc.) if you haven’t already. Many AI projects falter due to data silos or poor data quality. Invest in data governance and possibly hire data engineers or a Chief Data Officer if needed to oversee this strategic asset. On the technology side, continue your cloud journey; cloud gives you the scalability and access to AI services (“intelligence on tap”) that on-prem infrastructure might not. Evaluate your network and hardware needs too – some AI use cases might require edge computing or specialized hardware (GPUs, etc.). Simultaneously, double-down on cybersecurity and compliance. Implement robust identity and access management (so the right people – or AI agents – access the right data), encryption, and monitoring. If you’re leveraging third-party AI, scrutinize their security measures and incorporate them into your vendor risk management. Also update your policies: for instance, if employees are using tools like ChatGPT, provide guidelines on what data can or cannot be shared with such tools. Practical tip: Create an “AI governance checklist” for projects – covering data privacy, model bias, regulatory compliance (like GDPR, etc.), and security. This ensures each initiative is vetted properly. By fortifying your technical and security backbone, you create a reliable platform on which the flashy AI stuff can actually deliver. In my experience, projects that had early IT governance involvement went far smoother than those where governance was an afterthought.

• Start Small, Then Scale – and Learn at Each Step

It’s tempting to launch a grand, enterprise-wide AI program, but often a better approach is iterative. Identify a pilot project or two that are manageable in scope but significant in impact. Maybe it’s automating a specific manual process, or deploying a chatbot for a particular service line. Use these pilots to learn what works and what doesn’t in your context. Measure the outcomes and gather feedback from users. Then refine and expand. This agile approach lets you build success stories that can be used to get buy-in for larger rollouts. It also helps your organization build confidence and skills incrementally. When scaling up, think about change management – for a broader rollout, have a structured plan for training, communication, and support. One technique I found effective is a phased rollout: e.g., introduce an AI tool to one department first, iron out kinks, then progressively deploy to others. Treat those early departments as champions who can share their testimonials. Meanwhile, keep your eye on the strategic goal: ensure each small project is a stepping stone in your larger roadmap (as opposed to random experiments). This balance of quick wins and strategic direction is key. It’s much like climbing a mountain – you set the summit as the goal (vision), but you climb it basecamp by basecamp, acclimatizing and learning at each stage.

• Foster Collaboration Between IT and Business – Co-create the Future

Don’t let AI adoption be an “IT project” that is pushed onto the business. Instead, involve business leaders and subject matter experts from day one. Co-create solutions with them. If you’re developing an AI solution for, say, supply chain optimization, have your operations and supply chain teams deeply engaged in specifying requirements, testing outputs, and training the AI. This not only results in a better product but also secures advocates in the business who will champion the new solution to their peers. I recommend establishing interdisciplinary teams for each major AI initiative – pairing data scientists or AI engineers with business process owners. Also, share ownership of outcomes. For example, if the goal is to improve customer retention via an AI recommendation engine, make it a joint KPI between IT and the sales/marketing department. When success is shared, alignment is stronger. Regularly update all stakeholders (through steering committees or reports) about progress, hurdles, and results. This transparency keeps everyone invested. In my experience, such collaboration also surfaces creative ideas that IT alone might miss, and it roots the AI work in real-world practicality. In short, make AI a team sport across your organization.

• Lead with Vision, Empathy, and Adaptability

Lastly, a piece of advice about you – the leader. In times of great change, people look to leaders for direction and confidence. It’s important to articulate a compelling vision (why becoming a Frontier Firm will propel your organization forward) and to do so with genuine enthusiasm. Equally, lead with empathy – acknowledge the anxiety or skepticism that some may feel, and make it safe for them to voice concerns. Show that you are listening and adapting based on feedback. And be adaptable yourself. The AI landscape in 2025 and beyond is fast-moving; not every decision will be perfect, and new information might require course corrections. Be transparent about this. I often say to my teams:

“We’re charting new territory together; we’ll stay flexible and figure it out as we go.”

This builds a sense of camaraderie in facing the unknown. One practical habit: Schedule periodic retrospectives for your AI initiatives – what’s going well, what should we do differently? Include your team in these reflections. It reinforces that continuous improvement mindset. Finally, celebrate successes – even the small ones. Recognizing a team’s effort in successfully deploying a new AI feature or achieving a milestone goes a long way in keeping morale high and momentum going. Leadership in the Frontier Firm era is as much about inspiring and enabling people as it is about making tech decisions.

By following the above advice, IT leaders can serve as effective trail guides into this new territory of AI-powered business. We have the opportunity – and responsibility – to steer our companies wisely so that technology truly serves our strategy and our people. It’s an exciting road ahead, with challenges to be sure, but also immense potential for those who lead thoughtfully.

Final Thoughts: IT Leadership in the Frontier Firm Era

We stand at the frontier of a new era. Just as the first companies to harness electricity or the internet gained huge advantages, today’s Frontier Firms will be tomorrow’s trailblazers in the AI age. Becoming one of those organizations won’t happen by accident – it requires bold leadership. As an IT leader who has lived through waves of change, I can confidently say that this moment demands us to stretch beyond our comfort zones. We must be visionary – imagining what our businesses could look like when AI is infused into every process – and we must be courageous in pursuing that vision. There will be skeptics, and there will be obstacles. But it falls to us to make the case that the status quo is not an option when the competitive landscape is being redrawn by AI-native upstarts and forward-thinking incumbents.

The Frontier Firm era is not just about technology; it’s about redefining how we work and what our organizations value. It’s about:

Empowering people with tools that amplify their ingenuity
Freeing them from drudgery
Faster learning, experimentation — and yes, sometimes failure — but failing forward
Building companies that are more agile, creative, and resilient

To lead in this era, we have to marry optimism with pragmatism: embrace the art of the possible that AI brings, while carefully guiding our people and safeguarding our principles.

I encourage you, as leaders, to foster a sense of mission around this transformation. Rally your teams with the promise that by embracing AI, you’re not only future-proofing the organization, you’re also creating opportunities for each person to grow and focus on more fulfilling work. Share stories of positive change – whether it’s an employee who used an AI tool to solve a problem in minutes that used to take days, or a customer who received better service thanks to an AI-enhanced process. These narratives build momentum and buy-in.

Also, recognize that courage in leadership sometimes means saying yes to change before you have all the answers. It’s okay to move forward with imperfect information; in fact, in fast-moving domains, waiting for 100% certainty means you’re already late. Of course, manage risks wisely (as we discussed, governance is key), but don’t let fear paralyze innovation. Create a culture where calculated risks in the name of progress are rewarded. Your confidence will trickle down – when teams see leaders take bold yet thoughtful action, it inspires them to do the same.

As I reflect on my own journey, I’m struck by how much we’ve accomplished in just the past few years – things that once sounded like science fiction are now pilot projects on my desk. It reinforces my belief that the only real limit is leadership vision. The technology is largely ready; it’s the human decision to embrace it and reinvent our organizations that will determine who leads in the next decade. Microsoft’s Work Trend Index put it well:

“The time to act is now. The question for every leader and employee is: how will you adapt?”

This is a call I take to heart, and I hope you do too.

In closing, leading IT into the Frontier Firm era is the challenge and opportunity of our time. It asks us to be futurists and pragmatists, dreamers and disciplinarians, teachers and learners — all at once. But that’s what makes this journey worthwhile. I’m excited for the road ahead – for the innovations we’ll implement, the problems we’ll solve, and the new heights our teams and businesses will reach. Let’s lead with vision and courage, so that when history looks back on this era, our organizations will be remembered among those who stepped boldly into the frontier and thrived.

Sources

Microsoft Work Trend Index 2025 – Annual Report: “2025: The Year the Frontier Firm Is Born”
Stanford University (HAI) – AI Index Report 2025: Key Trends
McKinsey – “AI in the Workplace, 2025” Survey Highlights
Personal leadership experiences (T. Kilga) in digital transformation, AI adoption, cybersecurity (ISO 27001), and cloud migration.

Glossary: Key Terms in the Frontier Firm Era

AI Agent
An autonomous or semi-autonomous software program that can perform tasks, make decisions, or assist users based on AI models. In the Frontier Firm, agents are digital coworkers that augment human teams. Agent Boss
A new role every employee takes on — supervising, guiding, and collaborating with AI agents. Instead of doing all the work themselves, employees learn to delegate tasks to intelligent systems. Frontier Firm
A term coined by Microsoft to describe a company that fully integrates AI into its operations. These firms leverage AI at scale, combine human and digital workforces, and operate with speed, agility, and intelligence. Human–Agent Team
A collaborative working model where humans and AI agents function together, each contributing their strengths. These teams can evolve from basic assistantship to full task automation with human oversight. Intelligence on Tap
A concept where AI capabilities — like language processing, prediction, or analysis — are accessed on-demand via APIs or cloud services, similar to a utility like electricity. Prompt Engineering
The skill of crafting precise inputs (prompts) for generative AI models to produce desired outputs. This is a core competency in working effectively with large language models. AI Governance
The set of policies, practices, and frameworks that guide the ethical, secure, and compliant use of AI technologies within an organization. Digital Colleague
A metaphor for AI agents in a team. They don’t just assist — they can act, collaborate, and take responsibility for tasks, reshaping how teams operate. Work Chart
A future-forward alternative to the traditional org chart, focused not on hierarchies, but on "jobs to be done" — incorporating both human and AI roles. ISO 27001
An international standard for information security management systems (ISMS). It helps organizations manage data security and is foundational for AI-readiness in regulated sectors. AI Readiness
An organization’s preparedness to adopt and scale AI technologies — including the right infrastructure, data governance, skills, and culture.