Jacob Morgan

We tried to make work happier—and ended up making it hollow. Cultures softened. Leaders hesitated. Performance drifted. After 100+ CHRO interviews, I wrote The 8 Laws of Employee Experience—a blueprint to bring strength and humanity back to work. Learn more about it here: linkedin.com/pulse/my-new-b…

Worth reading to understand how one of the world’s ai leaders is thinking through the potential impacts (good and bad) over the coming years.

This Brown University story is easy to frame as a student cheating story, but I think it points to something much bigger. Inside Higher Ed reported that professor Roberto Serrano suspected many students used AI on a take-home midterm after the class average hit 96%. He later moved the final exam in person, where the average reportedly dropped to 48.6%, with 18 students dropping the class and 19 failing. The real issue is that AI is breaking a lot of the old ways we measure ability. A take-home exam, a writing assignment, a cover letter, a work sample, even parts of a performance review… all of these were built for a world where producing the answer was strong evidence that a person knew the material and that assumption is disappearing. The future of work version of this is going to show up everywhere. Hiring. Promotions. Learning and development. Manager feedback. Employee productivity. If AI can help someone produce polished work, then leaders need better ways to understand what people actually know, how they think, where their judgment is strong, and where they still need help. The answer can’t just be “ban AI.” That won’t work. We have to redesign assessment around the things that are harder to fake: reasoning, judgment, live problem-solving, explanation, collaboration, and the ability to apply knowledge in context. I talked about this on the latest episode of Future Ready Today Podcast.

This Brown University story is easy to frame as a student cheating story, but I think it points to something much bigger. Inside Higher Ed reported that professor Roberto Serrano suspected many students used AI on a take-home midterm after the class average hit 96%. He later moved the final exam in person, where the average reportedly dropped to 48.6%, with 18 students dropping the class and 19 failing. The real issue is that AI is breaking a lot of the old ways we measure ability. A take-home exam, a writing assignment, a cover letter, a work sample, even parts of a performance review... all of these were built for a world where producing the answer was strong evidence that a person knew the material, and that assumption is disappearing. The future of work version of this is going to show up everywhere. Hiring. Promotions. Learning and development. Manager feedback. Employee productivity. If AI can help someone produce polished work, then leaders need better ways to understand what people actually know, how they think, where their judgment is strong, and where they still need help. The answer can’t just be “ban AI.” That won’t work. We have to redesign assessment around the things that are harder to fake: reasoning, judgment, live problem-solving, explanation, collaboration, and the ability to apply knowledge in context. I talked about this an episode of Future Ready Today Podcast.

The most interesting part of OpenAI’s GPT-5.6 release isn’t the names Sol, Terra, and Luna. It’s the message behind them that not every job needs the most powerful model. Some work needs depth. Some work needs speed. Some work just needs to be handled cheaply and reliably at scale. That seems obvious, but a lot of companies still treat AI as if the answer is always “use the best model available.” That’s going to get expensive fast. The next phase of AI adoption won’t be about giving everyone access and hoping they figure it out. It will be about making better choices. Which model fits the work? Where is accuracy worth paying for? Where is a cheaper model good enough? Where should a human stay involved? You don’t send every task to the genius who charges a fortune. You match the tool to the work. That’s where AI is headed: less about access, more about judgment.

For a while, the AI conversation has mostly been dominated by the usual names in tech. But one of the more interesting signals now is that AI leadership is starting to show up in places many people don’t expect. Some of the companies getting real traction aren’t winning because they build the models. They’re winning because they’re figuring out how to use AI inside the business in a way that actually improves performance. It suggests AI advantage is becoming less about who can generate the most headlines and more about who can translate the technology into better operations, better decisions, better customer experiences, and better execution. In other words, this is starting to look like a management story as much as a technology story. Competitive advantage won’t come from saying you have an AI strategy. It will come from making AI useful in the flow of work and tying it to results that matter. The companies ahead right now seem to understand that. Curious what stands out most to you from the list.

Most leaders are spending a lot of time talking about what AI can do, but not nearly enough time understanding how it actually works. The gap between using AI and understanding AI is where a lot of bad decisions get made. It’s where companies overpay for the wrong models, trust outputs they should verify, deploy agents without guardrails, confuse confidence with accuracy, and mistake a fluent answer for a true one. The most important thing to understand is also the easiest to overlook: AI predicts. It doesn’t know. Tools like ChatGPT and Claude generate answers one token at a time based on patterns learned during training. This explains why these systems can be incredibly useful and dangerously wrong in the same afternoon. It explains hallucinations, context window limits, token costs, retrieval, agents, multimodal AI, and why a tool can sound certain even when it has no idea whether the answer is true. This is not a technical issue reserved for engineers. It’s a leadership issue. If AI is going to touch customers, employees, legal work, financial decisions, strategy, content, or operations, leaders need to understand where the machine is strong and where it needs structure around it. The goal is to deploy it with enough understanding that you know when to trust it, when to verify it, when to route it to a cheaper model, when to ground it in real sources, and when to keep a human firmly in the loop. I wrote a deep dive on the truth about how AI works and what every leader needs to understand before deploying it at scale. AI predicts. It doesn’t know. Deploy accordingly. Read the full article here: greatleadership.substack.com/p/inside-the-m…

Not every problem young workers are facing right now is because of AI. That’s what stood out to me from this Bloomberg piece. AI is changing entry-level work, but the bigger issue may be much simpler. Companies just aren’t hiring as much. That matters because entry-level jobs are where people learn how work actually works. How to deal with a manager. How to handle feedback. How to work with clients. How to miss the mark, fix it, and get better. None of that really happens in a classroom or a training module. When companies pull back on hiring young workers, they may save money in the short term, but they also weaken the talent pipeline they’ll need later. You can’t complain about a lack of experienced talent if nobody is willing to give people a place to start.

As someone who lives in California, this caught my attention. California is making Claude available across state agencies, cities, and counties through a new partnership with Anthropic. In plain English, AI is starting to move out of tech company demos and into the government services people actually use. That’s a big deal. The question is not just, “Can California use AI?” Of course it can. The better question is whether it can use AI in a way that is useful, responsible, and trusted. If AI is going to show up in public services, healthcare, DMV operations, cybersecurity, and citizen support, people need to know how it’s being used, where humans are involved, and how decisions are being made. This is where a lot of organizations miss the mark. They think adoption means buying the tool and giving people access. But the real work is training, governance, workflow redesign, transparency, and making sure the technology actually improves the experience. When AI moves into systems people rely on every day, the challenge is no longer just adoption. It’s earning trust.

There’s a conversation happening around return-to-office and leadership that I think needs a little more nuance. It’s easy to say, “Leaders who want people back in the office are just narcissists.” That might make for a good headline, but it doesn’t really help us understand what’s going on. The reality is that many people who end up in senior leadership roles tend to have traits like confidence, ambition, boldness, vision, and a willingness to be seen. Some of those traits can overlap with moderate levels of narcissism. In the right amount, they can help someone step up, make tough calls, and rally people around a direction. Too much, of course, becomes a problem. So when we study senior leaders and find higher levels of narcissistic traits, we have to be careful. It doesn’t automatically mean narcissism is causing every decision they make, including return-to-office preferences. It may simply mean we’re looking at a group of people who were selected, promoted, and rewarded partly because of traits that show up in many areas of leadership. This is why leadership is complicated. The same trait that helps someone become a leader can also become the thing that hurts their people if it’s not balanced with self-awareness, humility, and empathy. The question isn’t whether leaders should have confidence. They should. The question is whether that confidence is being used to serve the organization and its people, or just to protect the leader’s own preferences. I talk more about this in today’s episode of Future Ready Leadership, including what the research actually says about narcissism, leadership, and return-to-office. Listen here: podcasts.apple.com/us/podcast/the…

Oracle reportedly cut 21,000 jobs over the past year while pushing deeper into AI and cloud. A lot of people will look at this and say, “See, AI is coming for jobs.” Maybe. But I think the bigger story is that AI is forcing companies to confront the work they should have redesigned years ago. Most organizations are full of duplicated effort, slow processes, outdated roles, bloated approval chains, and teams that were built for a different era. AI didn’t create all of that. It just makes it harder to ignore. The danger is when leaders treat AI as a shortcut for cutting costs instead of a chance to rebuild work in a smarter way. Layoffs might improve a spreadsheet in the short term, but they don’t automatically create a more innovative company, a stronger culture, or a better employee experience. You can reduce headcount and still keep the same broken processes, the same bad management habits, and the same culture that slows everything down. The real question for leaders is not, “How many jobs can AI replace?” It’s, “How do we design a company where people and technology each do what they are best at?” That means looking at skills instead of job titles, removing work that shouldn’t exist anymore, helping people move into new areas before they are displaced, and giving employees clarity instead of vague corporate language about transformation. AI will change jobs. No question. Some roles will go away and new ones will appear. But leadership will determine whether that change feels like panic and cost-cutting, or like a thoughtful redesign of how the organization actually works. This is exactly the kind of conversation we’re having inside Future of Work Leaders, my private community for 40+ CHROs and senior people leaders focused on what’s next for work, leadership, and employee experience. Learn more or request an invite at futureofworkleaders.com.

Every company is trying to get better at prompt engineering. Better prompts, better outputs, better workflows, better AI agents. That matters, but it’s only half the conversation. The bigger leadership question is this: how do we get better thinking from the humans? As AI becomes more embedded in the way we work, one of the biggest risks is not that people will use it too much. It’s that human judgment will quietly go unused. Meetings get faster. Memos get cleaner. Strategy decks look more polished. But fewer people stop to ask: What did the AI miss? What did it get technically right but contextually wrong? What human experience, intuition, or moral judgment is being left out? That’s why I’ve been thinking about a concept I call Human Prompting. If prompt engineering is the art of getting better outputs from machines, Human Prompting is the art of getting better thinking from people. It means asking questions that activate the things AI cannot do: disagree when the room is wrong, use lived experience, make ethical judgments, tell the uncomfortable truth, bring real empathy, take ownership, and slow down when speed creates bad decisions. A few questions every leader should be asking in AI-enabled work: Where did you disagree with what AI produced? What did the AI get technically right but contextually wrong? What are we optimizing for that we actually shouldn’t be optimizing for? Who gets hurt if we’re wrong about this? What are we treating as an abstraction that is actually someone’s livelihood? What would a 20-year veteran know that the model would never know? AI will make organizations faster. That does not automatically make them wiser. The companies that win won’t simply be the ones with the best AI tools. Most organizations will have access to similar technology. The real advantage will belong to the companies that keep human judgment alive inside AI-powered decisions. This comes from a paid piece I published on Future Ready Leadership. Read full version here: greatleadership.substack.com/p/human-prompt…

AI’s free trial is ending. For the past two years, many companies treated AI like an open-ended experiment: buy the licenses, give people access, encourage usage, and assume value would eventually show up. Fortune reported that companies are now paying closer attention to AI spending, ROI, and whether employees are using expensive models for work that may not need them. That shift is overdue. The useful question is not, “Are people using AI?” It’s, “Are they using the right AI for the right work?” Not every task needs the most powerful model. Not every employee needs every tool. Not every workflow should be automated just because it can be. Usage without discipline becomes tool sprawl, confusion, and wasted spend. Eventually finance, IT, legal, HR, and business leaders all start asking the same question: what are we actually getting from this? The companies that win with AI will build a real operating model around it: where it should be used, where it should not be used, which models are good enough, when humans stay in the loop, how value gets measured, and who owns the cost. I talked about this on yesterday’s episode of Future Ready Today. Listen here: podcasts.apple.com/us/podcast/ai-… #AI #FutureOfWork #Leadership #Workforce #EmployeeExperience #FutureReadyToday

Most AI conversations focus on tools, productivity, and automation. But the conversation leaders need to be having is much bigger: How do you build trust when work is changing this quickly? How do you keep teams aligned when uncertainty is the new normal? How do you lead in a world where technology is moving faster than culture? That’s what we’ll be exploring on June 25 at 10 AM PST inside Future of Work Leaders, my private CHRO community. Our 40+ CHRO members will be joined by Patrick Lencioni, Founder & President of The Table Group and bestselling author of The Five Dysfunctions of a Team, for a private discussion on how to lead in the age of AI. Patrick has spent his career helping leaders build healthier teams and organizations, which is exactly the conversation we need right now. AI may change how work gets done, but leadership, trust, clarity, accountability, and culture are what determine whether organizations actually move forward. This is why we built Future of Work Leaders: a private, high-level community for senior people leaders who want to go beyond traditional HR conversations and focus on what’s next in leadership, talent, culture, AI, employee experience, and the future of work. We meet virtually every month and in person a few times a year. No vendors. No fluff. Just real conversations with the people leading this work inside major organizations. If you’re a CHRO or CPO and want to be part of these conversations, request an invite here: futureofworkleaders.com #FutureOfWork #CHRO #Leadership #AI #FutureOfWorkLeaders

China may be showing us what the AI workforce transition actually looks like. Not always dramatic layoff announcements. Not always a CEO standing on stage saying AI replaced thousands of jobs. Sometimes it is quieter than that: fewer roles backfilled, smaller graduate hiring classes, contractors being let go, teams being asked to do more with AI, and headcount slowly shrinking in the background. Reuters reported that some Chinese companies are adopting AI while quietly reducing hiring and headcount as Beijing pushes its “AI Plus” plan. That should get every leader’s attention because this is much harder to manage than a traditional layoff story. When jobs disappear through attrition, hiring slowdowns, and role redesign, there may be no big announcement for employees to react to. But people still notice. They notice fewer openings. They notice work changing. They notice when the company talks about AI productivity but avoids the workforce conversation. This is where AI adoption becomes a trust issue. If leaders treat AI only as a productivity lever, employees will assume the workforce plan is hidden. The better approach is to be clear about what work is changing, what skills will matter, where people will be redeployed, and how the company is going to support the transition. This is also why my conversation with Kelle Fontenot, Chief Digital Officer at KPMG, stood out. KPMG is not treating AI as a simple tool rollout. They are thinking through governance, adoption, digital teammates, workforce strategy, and how to help people use AI in the actual flow of work. That is the level of seriousness more companies need. AI adoption without a workforce plan becomes a trust problem. I talked about this broader shift with Kelle on Future Ready Leadership: podcasts.apple.com/us/podcast/how… Source: Reuters, June 2026 #AI #FutureOfWork #Leadership #Workforce #HR #EmployeeExperience

Everyone wants AI to eliminate entry-level work. Very few leaders are asking what happens after that. Junior employees don’t become senior by reading a policy document or watching a training video. They become senior by doing the messy work. The research. The first drafts. The analysis. The scheduling. The customer questions. The reporting. The revisions. The mistakes. That work can look boring from the outside, but it is often where judgment gets built. This is the apprenticeship problem AI is creating. If we automate the bottom of the career ladder, we may save time today and create an expertise shortage tomorrow. Junior analysts learn by building the model. Junior lawyers learn by reviewing documents. Junior engineers learn by debugging. Junior HR partners learn by handling the operational issues before they become strategic. AI can absolutely remove low-value work, but companies need to be careful about confusing “low-value” with “developmental.” Some work is inefficient and should go away. Some work is how people learn. The question for leaders is not just, “What can AI automate?” The better question is, “If AI does this work, how will people build the judgment they used to get from doing it?” The future-of-work issue isn’t just job displacement. It’s expertise development. The companies that figure this out will not just use AI to cut work. They’ll redesign how people learn, grow, and become the next generation of experts. I'm going deeper on this on my Substack:  greatleadership.substack.com/p/how-to-creat…

Another great leader joins Future of Work Leaders. Please join me in welcoming Véronique Subileau, Senior Vice President, Global Human Resources at UGI Corporation, to the community. Future of Work Leaders is now made up of more than 40 CHROs and senior HR executives who are focused on where work is going next, not traditional HR for the sake of HR. We meet virtually every month and a few times a year in person to talk about the real issues shaping organizations: culture, leadership, AI, talent, employee experience, workforce expectations, and the changing role of the CHRO. The value of this group comes from the people in the room, and I’m excited to have Véronique be part of it. Welcome, Véronique. Learn more and request an invite here: futureofworkleaders.com

Everyone keeps calling SpaceX a rocket company. I’m not sure that’s accurate anymore. The IPO is being treated like a huge moment for rockets, Starlink, and Elon Musk, and of course it is. But the more interesting part is what’s happening underneath the headline. Starlink appears to be the business making money. Rockets are still expensive. And now xAI is pulling a massive amount of capital into compute, data centers, and the race to build better models. So what went public wasn’t really one clean business. It was three very different bets wrapped together: satellite internet, reusable space infrastructure, and AI. That’s why the valuation caught my attention. Investors aren’t just pricing what SpaceX is today. They’re pricing a version of the future where Starlink keeps growing, Starship works at scale, and xAI becomes a serious AI infrastructure player. Maybe that happens. Elon Musk has made a career out of doing things people said couldn’t be done. But when a company is valued on a story that big, the story has to keep getting validated. The bigger signal for me is that AI infrastructure is moving from private-market patience into public-market pressure. Venture capital can tolerate long timelines and huge losses. Public markets eventually want numbers: revenue, margins, and a believable path to profitability. That’s where this gets interesting for leaders. AI is no longer just a tool conversation. It’s becoming a capital markets conversation, a business model conversation, and an infrastructure conversation. SpaceX may turn out to be one of the defining companies of the next decade. But this IPO is also a reminder that the AI race is getting more expensive, more public, and more exposed to pressure. I discussed this in detail on today's episode of The Future Ready Podcast: podcasts.apple.com/us/podcast/spa… #FutureOfWork #AI #Leadership #SpaceX #FutureReadyToday

AI is quickly becoming part of how work gets done, but many companies still have no idea who should use it, how much they should use, or what the business should get back. @Uber offers a useful example of what happens when AI adoption scales quickly. The company reportedly reached 95% monthly adoption of AI coding assistants, burned through its planned 2026 AI budget in just four months, and then began capping token usage by role and function. The lesson is not to slow AI adoption. It is that adoption without governance eventually turns into restriction. Early on, AI feels like a simple productivity upgrade. Give employees access, encourage experimentation, and celebrate usage. That works when the goal is learning. But as adoption spreads, the questions change: • Where is AI creating value? • Which roles need the most access? • Who owns the budget? • Who is accountable for outcomes? These questions matter because AI usage is not equal. An engineering team accelerating product development creates different value than a team experimenting without a clear business case. That is why leaders need three decision gates before scaling AI. 1. Value: Prove a use case delivers measurable results before expanding access. If AI improves customer support resolution times or reduces sales prep work, measure it. 2. Ownership: Assign accountability for budget, usage, vendor management, and performance. Without ownership, costs and responsibility become fragmented. 3. Access: Match AI access to business need. High-impact teams may require advanced capabilities, while others may only need AI embedded in existing workflows. This is not about being conservative with AI. It is about being disciplined. Companies should absolutely experiment. But experimentation and governance need to mature together. Otherwise, the pattern is predictable: access expands, usage accelerates, spending rises, visibility declines, and leadership responds with limits after costs are already out of control. The better approach is simple: identify where AI creates measurable value, establish accountability, and align access with business priorities. AI should scale with purpose, not just enthusiasm. That is how organizations move from AI experimentation to AI discipline.

Hybrid work didn’t fail. Vague hybrid did. Dropbox recently called hybrid “the worst of both worlds,” and I get why. A study from Drake University researchers Radostina Purvanova and Alanah Mitchell followed employees at three large financial services companies from 2022 to 2025. The companies weren’t named, but one went fully back to the office, one stayed fully remote, and one used a hybrid model. Hybrid looked like the winner at first. Approval went from 44% in 2022 to 63% in 2025. But only about half of the people who preferred hybrid in 2022 still preferred it three years later. The rest drifted toward either fully remote or fully in-office. That’s the part leaders should be paying attention to. Hybrid is usually sold as the reasonable compromise: employees get flexibility, leaders get office time, HR gets a policy that sounds balanced. But too many companies turned it into “come in sometimes and figure it out on your own.” That’s how people end up commuting to the office to sit on Zoom, teams show up on different days, managers don’t know who will be where, and employees wonder why they came in at all. Hybrid can work, but it needs design. Shared office days. Clear team norms. A reason to be together. A better match between the work and the location. Deep work at home makes sense. Mentoring, problem-solving, relationship building, and collaboration often make more sense in person. The mistake is pretending flexibility means no structure. This is also the mistake many companies are making with AI. They hand people tools and subscriptions, but don’t redesign the work. Then they’re surprised when the results are messy. Hybrid and AI have the same lesson: new ways of working don’t fix bad operating models. They expose them. I talked about this in the previous episode of Future Ready Today: podcasts.apple.com/us/podcast/why…

A future ready organization is built on principles that shape everything from how you lead to how you develop your people to how you approach technology.