Your AI Wake Up Call
We’re not just witnessing another tech wave.
We’re entering an entirely new chapter for humanity.
This is not about incremental improvements like the next iPhone upgrade or the next operating system. It’s about a fundamental shift in how intelligence operates on our planet. Just as the Enlightenment moved us from faith-driven to reason-driven thinking, we’re now shifting from human reasoning to the reasoning of nonhuman intelligences.
What Most Leaders Get Wrong
There are a couple of areas where many leaders are starting to fall behind:
First, they dramatically underestimate the speed of AI advancement. The capabilities we’re seeing today would have seemed impossible just two years ago, yet many leaders are still planning as if change happens gradually. This week, I interviewed a former design lead at IDEO, and he said about 15% of leaders he speaks with reject using AI. Admirable in some ways, but also a very dangerous thing to do.
Second, they miss the full scope of transformation. Making your customer service chatbot slightly better is not what this transformation is about. At least not anymore. And I get it. I also consistently work on improving our chatbots within the Strategy & Consulting AI system. If you were a member from the time we launched, you know that we rebuilt that system a few times already, from the ground up. Every time it is getting better and better. And a lot more updates are coming. Chatbots are still and will likely be for a long time helpful for members. Here is a message we just received from one of the great new members of our community:
But chatbots are just a small portion of what we do AI-related. And I recommend that if you were just focusing on chatbots, it's time to start paying attention to how you can integrate AI more broadly. Just focusing on chatbots is no longer nearly enough. It’s about reimagining how work gets done at every level. Basically, take a look at your business and redesign it. Can you put AI and technology at the core of your organization?
Third, they rely on outdated strategies. If you’re using AI strategies built even two years ago, you’re already behind. Most leaders are not prepared and better get organized, for both the opportunities and the risks.
The Consensus on Core Technologies
Across top AI teams, there’s now alignment on four technologies that will transform everything. Four forces driving transformation are reasoning at scale, agents coordinating end-to-end tasks, memory that compounds learning, and multimodal systems that blend text, vision, and sound.
These are expected to transform nearly everything in the next three to six years. That timeline is already influencing where talent flows, how capital is allocated, and the direction of policy debates.
The Agentic Revolution
When I renovated my home in California, I managed endless permits, contractors, and change orders. It was exhausting, expensive, and full of inefficiencies and unnecessary stress.
Now imagine if, instead, linked AI agents had handled the entire process:
One agent identifies contractors to hire, interviews them, evaluates them, and reviews and negotiates contracts. And they would do a better job than most humans because most of us will only do 1-2 full renovations in our lifetime, if any, so we just don't know what we don't know. The agent will know to put penalties for cost overruns and time overruns. The agent will know to specify the quality of work that is required before payment will be made, because contractors can do a luxury house renovation and think that crooked or cracked tiles are an appropriate deliverable.
Another navigates permits codes.
Design agents create optimized plans for renovation.
Another agent finds relevant materials to use that are aligned with design and budgets.
Structural agents ensure safety compliance.
Procurement agents hire additional specialized contractors and coordinate schedules.
Legal agents handle disputes and contracts.
Quality control agent reviews the deliverables and flags issues to resolve.
We are getting closer and closer to the time when this is no longer science fiction. This kind of chained workflow automation will soon be the norm, at least to some degree. Whether you’re in consulting, banking, pharmaceuticals, healthcare, or government, every process is up for reinvention. And it’s important to start that reinvention work now.
The Runaway Intelligence Problem
Here’s the elephant in the room: self-reinforcing intelligence loops. AI systems that learn, test, and upgrade themselves faster than traditional governance can follow.
Nobody, not policymakers, not researchers, has a robust approach for aligning or containing this. Current solutions are partial at best.
For businesses, this means you can’t wait for perfect regulatory frameworks. You need to build governance and risk management infrastructure immediately, or else you’ll be making decisions without a clear view of what’s happening while AI capabilities advance at machine speed.
This week, I interviewed Bree Groff, who has helped organizations like Microsoft, Google, Target, and Hilton. I asked her about her thoughts on AI advancement. She told me:
“If the AI is a train speeding up behind you, don’t try to outrun it. Step off the tracks.”
You need to understand what AI can do better than you, step away from those tasks, and focus on work AI can’t do yet, while leveraging what AI can do.
The New Industrial Moat
Many people assume the race in AI is about talent or algorithms only. But the deeper reality is the new moat is compute.
GPUs and energy are the new oil fields of the digital economy. Without them, you’re not even in the game.
Think about how railroads once determined which cities did well, or how oil determined which countries had leverage in the 20th century. Compute capacity is playing that role now. Whoever controls access to large-scale processing power will anchor the next industrial platforms.
For leaders, this means compute is not an IT issue. It’s a boardroom issue. It’s not only about hiring more and better engineers. It’s about securing the infrastructure that makes all engineering possible.
And at the individual level, understanding this shift matters just as much. If you don’t see where the bottlenecks and leverage points are moving, you will not be able to make good decisions for yourself and your organization.
The Global AI Standards War
There's also a geopolitical dimension that affects every business. The West is pursuing closed, proprietary systems with controlled access and revenue-focused business models, but with limited geographic reach. Meanwhile, China is pursuing open-source dominance through open-weight models, global accessibility, state-backed development, and rapid international deployment.
If open-source models gain dominance in large parts of the world, technical leadership in the West may prove temporary as global standards shift away from Western platforms. For businesses, this means interoperability and policy engagement must become executive priorities, not IT afterthoughts. And for us at an individual level, it means we have to pay attention to what is happening in the world, not just within the borders of our country or in the West.
Your Action Plan
So what do you actually do? You can consider taking five concrete actions.
First, audit your workflows. Are your processes agent-ready, or bottlenecked by manual steps that will slow AI integration? At an individual level, that means looking at your workflows and how they should be adjusted and where AI and technology can be integrated.
A few days ago, I spoke to the former CEO of 2 Fortune 500 companies (Alcoa in the US and Siemens in Germany), Dr. Klaus Kleinfeld. He mentioned that it is important to not only test the tools and keep what works and discard what doesn’t, but also return to tools because some of them are improving very fast, and something that wasn’t useful 2 months ago can be a very useful tool today.
Second, build governance and risk frameworks now. Invest in oversight, stress-testing, and explainability for every AI deployment before they become regulatory requirements.
Third, if your organization requires it, partner with or build hardware capacity, don't just rely on talent or off-the-shelf tools.
Fourth, engage globally. Monitor and adapt to worldwide innovation, not just your home market trends.
Fifth, and one of the most important ones, think in weeks, not years. Prioritize rapid pilots, measure results, adjust quickly, and avoid trying to do everything perfectly. That is not something we can afford right now.
The Window Is Closing
If you’re leading a business, transformation is not a matter of future strategy; it’s the ground on which we compete right now. Whether the mainstream shift happens in three years, six years, or longer, the structures you build today (governance, capital allocation, and execution speed) will determine your trajectory.
We must take action before it feels comfortable. We have to move faster than feels natural. We have to learn to build while we learn. And whatever we do, we can’t let this window close on us. The companies and individuals that will do well in this next chapter for humanity will be those that act decisively, while others are still debating whether this is real or hoping it will take a long time for things to really change.
To elaborate a little more on actions to take for you individually, because that is what is likely most applicable to you, here are some things to consider:
Embrace “continuous reinvention.” The skills, workflows, and competitive factors that mattered five years ago are already shifting. Don’t wait for someone else to set the pace for you. You need to increase your own learning velocity.
Develop operational AI literacy. Know enough to distinguish what is important from what is not important, and to prioritize the most promising pilot projects or strategic experiments. And don't just prioritize it. Allocate time weekly to implement. Don’t delegate this understanding entirely to people on your team. AI is now a core leadership competency.
Build and lean on networks. I credit staying close to brilliant peers, finding mentorships, and collaborating as key to staying somewhat ahead. When we work with executive coaching clients, we always strategize how to have this powerful networking component in place (who, how, when, why). Just reading about these technologies is not enough.
Audit your personal value-add. As routine work is progressively automated, focus on the roles, problem-solving, and human creativity that AI cannot yet replicate. The entire StrategyTraining.com library is there to help you build such skills that AI will not be able to replicate for a long time: superior business judgment, advanced ability to predict which strategy will likely give you the biggest competitive and comparative advantage. AI is still very bad at predicting the future, and we are currently working on an important study for the SCRA that covers some of the biggest failures when it comes to AI's ability to predict and the significant cost and harm it already causes to businesses, individuals, and society.
Speaking of StrategyTraining.com resources, you want to continue developing deep skills in general problem-solving and the ability to influence people within the organization. Ask: “How am I adding value on top of AI, and how do I help my team do the same?”.
Prioritize “speed of iteration.” Run experiments, test prototypes, and adopt a bias for action. Mistakes of hesitation are the costliest right now.
AI transformation will not wait for when we feel comfortable with this new technology or traditional cycles. As leaders, the imperative is to keep learning, keep iterating. And, of course, our individual adaptability shapes organizational success and maximizes the chances that our skills will continue to be in demand.