Leave a comment

The company moved forward with layoffs and organizational changes, but internal evidence showed that the technology hadn’t produced the expected productivity. Technical disruptions increased, employees pushed back, and Meta canceled a planned second round of workforce reductions. By July, Mark Zuckerberg acknowledged that agent development had progressed more slowly than expected. A Reuters’ investigation showed what happened when expectations for AI agents moved faster than their demonstrated performance.

Leave a comment

I didn’t see Meta’s experience as an argument against AI agents. I saw it as evidence that companies had treated agents as technology deployments when they should’ve treated them as a new category of worker. Organizations gave agents assignments, system access, data, and decision-making authority. They didn’t always give them defined responsibilities, measurable standards, ongoing supervision, or structured performance reviews.

Deployment Isn’t the Finish Line

Traditional software either executes its programmed logic or produces an identifiable error. AI agents operate differently. They interpret objectives, select tools, make intermediate decisions, and adjust their actions as conditions change. Two executions of the same assignment can produce different results.

That variability makes uptime, speed, and token usage inadequate measures of success. An agent could remain available, respond quickly, and complete thousands of actions while still damaging the business. It might close support cases without resolving customers’ problems, generate sales leads that never converted, or produce code that increased the review burden on engineers.

The 2026 APEX-Agents benchmark evaluated agents across 480 realistic tasks in investment banking, management consulting, and corporate law. The agents worked across documents, spreadsheets, email, calendars, and other workplace tools. When the benchmark was released, even the strongest systems completed fewer than one-quarter of the professional assignments successfully on their first attempt.

A separate 2026 study, Towards a Science of AI Agent Reliability, evaluated 14 models using 12 measures across consistency, robustness, predictability, and safety. The researchers found that capability improvements had produced only modest gains in reliability. A model could perform well on average and still fail unpredictably across repeated runs.

I believe that distinction matters. Capability tells me what an agent could do under favorable conditions. Performance management tells me what it did consistently inside my business.

Every Agent Needs a Performance Contract

I would have started each deployment with a performance contract. That doesn’t mean an employment agreement. It means a precise definition of the agent’s job, the results it was expected to produce, the decisions it could make, and the conditions that required human intervention.

A customer-service agent, for instance, shouldn’t have been evaluated by the number of conversations it handled. I would’ve measured verified resolution, repeat contacts, escalation quality, customer satisfaction, policy compliance, and the cost per completed resolution. Those measures would’ve revealed whether the agent created value or just activity.

I also would have evaluated the path the agent took. An agent might reach the correct outcome after accessing unnecessary data, calling the wrong tools, repeating expensive actions, or bypassing an approval. The result could look successful while the process introduced risk. Performance management had to examine both the outcome and the conduct that produced it.

Agents Need Reviews and Consequences

Human performance management doesn’t end with an annual rating, and agent performance management shouldn’t end with a pre-deployment test. I would have reviewed agents continuously because models, data, tools, customer behavior, and business rules kept changing.

Those reviews would have compared actual results with defined thresholds. They would have identified recurring failure patterns, tested performance after system changes, and examined cases in which humans corrected or reversed an agent’s work. The findings could then guide better instructions, narrower permissions, additional training data, revised workflows, or a different model.

Agents need consequences. A high-performing agent could receive broader authority. An inconsistent one might require more approvals. An agent that violated safety or compliance limits might be suspended until the failure was understood. Autonomy should be earned and retained through demonstrated performance.

I expect AI agents to become an important part of how companies operate. That makes management more important, not less. The organizations that succeed won’t be the ones that deploy the most agents. They’ll be the ones that know which agents perform, which ones fail, why they fail, and when human judgment still produces the better result.

LEARN MORE

Leave a comment

Instead, AI has created a requirements crisis. The better AI becomes at producing software, the more important it becomes for people to define exactly what that software is supposed to do. Code is becoming easier to generate, while clarity is becoming harder to find.

Leave a comment

The need for clarity challenges how many companies approach AI-assisted development. They invest in coding tools because implementation appears to be the largest source of time and cost. Once AI accelerates implementation, however, the bottleneck doesn’t disappear. It moves into domain understanding, requirements, architecture, validation, integration, and maintenance.

When a human developer receives an ambiguous requirement, that person may stop, ask questions, challenge assumptions, or recognize that two business rules contradict each other. An AI agent can ask questions as well, but it can also fill the gaps with plausible assumptions and continue working. And that’s a problem. Because AI operates at machine speed, a small misunderstanding can negatively impact thousands of lines of code before anyone notices.

That’s where speed becomes dangerous. If a team misunderstands a requirement in a traditional development process, the error may appear in one feature developed over several weeks. With AI agents working across multiple tasks, the same misunderstanding can spread through source code, tests, documentation, database structures, and integrations in a matter of hours.

Requirements Debt

I think of the accumulated result as a requirements debt. Technical debt describes compromises embedded in software. Requirements debt describes ambiguity, missing context, conflicting rules, and untested assumptions embedded in the instructions used to create it. Once those weaknesses enter an AI-driven workflow, each new generation cycle can carry them forward.

There’s also the risk that an AI-generated system can satisfy the written specification and still fail to solve the real business problem. If the requirement says that a loan application should be approved when a particular score exceeds a threshold, the agent can implement that instruction with precision. It won’t know that a compliance rule requires a second review, that a data field has become unreliable, or that the policy applies only in certain states unless someone includes those conditions.

That’s why software quality is no longer limited to whether the code matches the specification. We also have to ask whether the specification matches the business, the customer, the law, and the operating environment. AI can help verify the first question. Human judgment remains essential for answering the others.

Requirements Are Part of the Production System

Companies have traditionally treated requirements as documents that developers need before coding begins. In an AI-driven environment, requirements become part of the production system. They need owners, version control, traceability, acceptance criteria, business-rule validation, and continuous revision. If an organization can’t explain a requirement in a form that people and machines can test, it isn’t ready to delegate the work to an autonomous agent.

This doesn’t mean every requirement needs to become a massive technical document. More documentation doesn’t guarantee more clarity. Teams need specifications that identify the intended outcome, relevant constraints, exceptions, data sources, security rules, acceptance criteria, and the person responsible for resolving conflicts. The goal isn’t to describe every keystroke. It’s to remove the uncertainty that can change the result.

The Need to Understand Intent

AI is also increasing the value of people who understand the business and can translate that knowledge into explicit decisions. Product managers, business analysts, architects, quality engineers, security professionals, and experienced developers aren’t becoming less important. Their work is moving closer to the center of software creation because they define the boundaries within which AI operates.

This changes the skills companies need from their engineering teams. Writing code still matters, but understanding intent matters more. The most valuable professionals increasingly know how to identify missing information, expose conflicting assumptions, define measurable outcomes, and determine what evidence will prove that a system works. They don’t just tell the AI what to build. They create the conditions under which its output can be trusted.

The future of software development doesn’t depend only on how much code AI can produce. It depends on whether people can define the right problem, specify the right outcome, and validate that what the machine builds is what the business needs.

AI has made code abundant. But clarity continues to remain scarce.

LEARN MORE

Leave a comment

Leave a comment

Instead of asking AI to generate a class or fix a bug, I gave it an objective. The system decomposed the work into dozens of smaller tasks, assigned those tasks to specialized AI agents, reviewed their output, requested revisions when necessary, and assembled the finished product. The code no longer came from a single large language model. It came from an organized team of AI specialists coordinated by what the industry now calls an Agent Manager.

Deciding How Work Gets Done

An AI agent performs a task. An Agent Manager decides which agent should perform that task, determines the order in which the work should occur, evaluates the quality of each result, and redirects work whenever something goes wrong. In many ways, it resembles an experienced engineering manager who coordinates a team of specialists rather than writing every line of code personally.

This wasn’t an isolated trend. Gartner predicted that by 2026, 40 percent of enterprise applications would incorporate task-specific AI agents, compared with fewer than 5 percent only a year earlier. At nearly the same time, Deloitte’s 2026 Global Software Industry Outlook concluded that autonomous AI agents were becoming central to software engineering, changing how organizations build products, manage engineering teams, and allocate technology investments.

Those reports pointed toward the same conclusion. As organizations deployed more AI agents, someone had to coordinate them. That responsibility increasingly belonged to the Agent Manager.

I found the implications fascinating because they represented a complete shift in the software development lifecycle. One agent gathered requirements. Another designed the architecture. A coding agent generated the implementation. Separate agents produced automated tests, scanned for security vulnerabilities, reviewed compliance requirements, updated documentation, and prepared deployment pipelines. The Agent Manager supervised every step. It tracked dependencies, evaluated progress, requested corrections, and determined when work was ready for human review.

The Role of Human Expertise

Instead of spending most of the day writing syntax, developers increasingly define objectives, establish constraints, review outcomes, and exercise judgment. As AI made software production dramatically faster, human expertise became concentrated where it always created the greatest value -- deciding whether the software actually solved the right business problem.

In July, Microsoft Research published The Rise of AI Coding Agents, an analysis of enterprise adoption across tens of thousands of software engineers using autonomous coding systems. The researchers found that engineers working with agentic coding tools merged approximately 24 percent more pull requests than expected. The paper also reached an important conclusion that received less attention. Organizations that achieved the greatest improvements weren’t necessarily using better AI models. They were better at integrating AI agents into structured engineering workflows and management practices.

The same month, researchers published Human Oversight in Agentic Software Development on arXiv. The study analyzed more than 25,000 AI-generated pull requests across 2,361 public GitHub repositories. The authors discovered that successful projects almost always relied on continuous human oversight throughout the development process. Fully autonomous software development remained rare. Human managers reviewed decisions, resolved conflicts between agents, validated outputs, and intervened whenever quality declined.

The Increasing Importance of Software Validation

AI didn’t eliminate software quality as a competitive advantage. It increased its importance. When generating code becomes inexpensive, validating that code becomes dramatically more valuable. The Agent Manager exists not to maximize code generation but to maximize confidence in the software that reached production.

Researchers from several universities published Lessons from a Twelve-Week Agentic Software Engineering Project in July 2026 after documenting a production software initiative built largely through frontier AI coding agents. Rather than celebrating autonomous development, the authors concluded that organizations needed stronger governance, repeatable engineering controls, and continuous supervision as AI systems became more capable. Greater autonomy increased the importance of management rather than reducing it.

For decades, software managers assigned work to developers. Tomorrow’s engineering leaders will increasingly assign work to intelligent agents. Their responsibilities will extend beyond sprint planning and code reviews. They’ll monitor agent performance, optimize task allocation, evaluate reasoning quality, manage operating costs, enforce security policies, and determine where human judgment should remain part of the workflow.

AI Governance and the Future

The engineering dashboard of the future won’t focus exclusively on developer productivity. It will also measure how effectively the Agent Manager coordinated specialized agents, how frequently human intervention became necessary, how reliably AI-generated software passed validation, and how efficiently computational resources were consumed.

That shift also explains why AI governance has become one of the fastest-growing areas of enterprise investment. Deloitte’s 2026 Global Software Industry Outlook identified governance, operational controls, and responsible AI oversight as critical factors separating successful AI deployments from expensive pilot projects that never reached production. The technology itself was only part of the equation. Sustainable value depended on disciplined management.

Software development has always been an exercise in coordination. The people involved are changing, but the need for leadership isn’t. Yesterday, engineering managers coordinated teams of developers. Tomorrow, they’ll coordinate teams of AI agents. The technology will continue advancing at an extraordinary pace, yet the organizations that thrive won’t be the ones generating the most code. They’ll be the ones managing autonomous who manage autonomous intelligence better than everyone else.

LEARN MORE

Leave a comment

Rather than handing employees another chatbot, Uber created what it calls “Agentic Pods,” small teams of AI engineers embedded within business functions such as finance, legal, marketing, human resources, and procurement. Their mission wasn’t to automate individual tasks. It was to observe how work actually happened, redesign entire workflows around AI, and build intelligent agents that could execute large portions of those workflows. The results were remarkable. Financial planning fell from 15 hours to 30 minutes. Financial reporting dropped from two days to 10 minutes. Marketing quality assurance shrank from two weeks to less than an hour.

Leave a comment

For the past twenty years, digital transformation has focused on helping people work better through technology. Organizations replaced paper with software, moved applications to the cloud, digitized customer interactions, and automated repetitive processes.

AI transformation changes the equation.

Instead of asking how technology can help people perform work, AI transformation asks which work should be performed by intelligent systems and which work should remain human. That distinction changes everything. It reshapes jobs, workflows, decision-making, and even the structure of the organization itself.

I’ve come to believe that successful AI transformation depends on three pillars. Ignore one of them, and the entire effort becomes unstable.

Pillar #1 … People

Companies don’t transform. People do. The biggest mistake I see is leaders treating AI as a technology project instead of a human one. They purchase enterprise licenses, deploy copilots, and expect productivity to improve automatically. It rarely does.

Employees need new skills. They need to understand not only how to use AI but how to supervise it, question it, validate it, and collaborate with it. Critical thinking becomes more valuable, not less. Judgment becomes a competitive advantage.

The luxury retailer Tapestry, the parent company of Coach and Kate Spade, offers an excellent example. Rather than limiting AI to the technology department, the company focused on changing how employees worked. It organized a companywide AI summit, created AI personas representing Gen Z shoppers to help teams understand customer preferences, introduced AI coaching for retail associates, and used AI-assisted design with 3D printing to reduce prototype development time by 75 percent. Yet executives repeatedly emphasized that the real transformation wasn’t the software. It was the culture. Human designers still made the final decisions. AI expanded creativity instead of replacing it.

That’s exactly what successful AI transformation looks like. People don’t become less important. They become more important because they’re responsible for directing increasingly capable intelligent systems.

Pillar #2 … Process

This is where AI transformation separates itself from digital transformation. Digital transformation improved existing processes. AI transformation redesigns them. Too many organizations ask how AI can automate today’s workflow. I think the better question is whether today’s workflow should exist at all.

Every business process should begin with three simple questions.

• Does this activity create value?

• Can AI perform part of the work?

• Should the process itself be redesigned before AI is introduced?

The organizations generating the greatest returns aren’t automating yesterday’s processes. They’re inventing entirely new ones.

McKinsey reinforced this conclusion in its 2026 research on enterprise AI adoption. The firm found that organizations create the greatest business value not by giving more employees access to AI, but by redesigning workflows, operating models, and organizational roles around what AI makes possible. Companies that merely automate existing processes tend to realize incremental gains. Companies that rethink how work flows through the enterprise create transformational results.

Pillar #3 … Trust

Every AI success story eventually reaches the same point. Can we trust the output? The faster AI becomes, the more important validation becomes.

Organizations need governance that establishes accountability, security, privacy, quality, and explainability. Every critical AI decision should have a human owner. Every model should be monitored. Every important output should be validated before it influences customers, finances, or operations.

I don’t view governance as bureaucracy. I view it as the foundation that allows organizations to scale AI with confidence. The software industry illustrates this perfectly.

AI can now generate thousands of lines of code in minutes. That doesn’t mean the software is secure, maintainable, compliant, or aligned with business requirements. In many development organizations, the bottleneck has already shifted. Writing code is no longer the slowest part of software development. Validating AI-generated code has become the scarce resource.

The same principle applies far beyond software engineering. The more work AI performs, the more valuable human judgment becomes.

AI Transformation Is Business Transformation

Looking back, I think we’ve been using the wrong comparison. AI transformation isn’t the next chapter of digital transformation. It’s the beginning of a new operating model.

• Digital transformation digitized work. AI transformation redistributes work between people and intelligent systems.

• Digital transformation focused on efficiency. AI transformation focuses on redesign.

• Digital transformation changed the tools employees used. AI transformation changes the role employees play.

That’s why I believe every successful AI transformation rests on three pillars. Develop people who know how to work alongside intelligent systems. Redesign processes instead of automating outdated ones. Build the trust that allows AI to operate at enterprise scale. Get those three pillars right, and AI stops being another technology investment.

It becomes the way the business creates value.

LEARN MORE

Leave a comment

Grigorev’s experience wasn’t an indictment of AI. It was a reminder that faster code generation doesn’t eliminate the need for human validation. If anything, it makes that validation more important than ever.

The Quality Risks

For the past two years, we’ve celebrated how AI accelerates software development. Developers are producing more code than ever before. Release cycles are shrinking. Engineering teams are delivering features at a pace that would’ve seemed impossible only a few years ago. Those gains are real, and I don’t believe we’re going back.

What concerns me is that we’ve become so focused on development speed that we’ve overlooked the growing quality risks accompanying it. AI has fundamentally changed how software is written, but it hasn’t changed the standards that determine whether software is ready for production.

Every line of generated code still has to be secure, maintainable, and aligned with the surrounding architecture. It also needs to handle unexpected inputs, integrate with existing systems, and support future enhancements that nobody has yet imagined. AI writes code remarkably fast. Determining whether that code actually belongs in production remains a human responsibility.

What If the Code Looks Correct?

In a lot of cases, AI-generated code compiles successfully, passes basic tests, follows accepted coding conventions, and appears every bit as polished as code written by an experienced developer. This creates a dangerous illusion. Hidden beneath that clean syntax may be flawed business logic, subtle security vulnerabilities, duplicated functionality, inefficient algorithms, or architectural decisions that increase technical debt.

Researchers continue to document this growing problem. A 2025 study published in ACM Transactions on Software Engineering and Methodology evaluated GitHub Copilot across a wide range of programming tasks and found that the correctness of its suggestions varied significantly depending on the problem being solved. The researchers concluded that developers cannot assume AI-generated code is reliable without careful review and verification.

Leave a comment

Another warning comes from Communications of the ACM, where researchers explained that large language models learn from enormous public code repositories containing both excellent engineering practices and insecure coding patterns. As a result, AI assistants can unknowingly reproduce vulnerabilities that already exist within their training data, making human review an essential safeguard rather than an optional step.

The AI Quality Gap

Developers who once reviewed a few hundred lines of manually written code now receive thousands of AI-generated lines in the same amount of time. Reviewing that much code with the same level of rigor is extremely difficult. The temptation is to trust the AI because the code looks professional and because productivity metrics suggest everything is moving in the right direction.

The front end of software development has accelerated dramatically. Requirements become working code almost immediately. Prototypes appear in hours instead of weeks. Feature velocity continues to climb. Yet the activities that determine software quality, including architecture reviews, security testing, integration validation, regression testing, and business acceptance, haven’t accelerated at the same pace.

This shift increases the value of experienced software engineers rather than reducing it. Writing code is becoming easier. Evaluating code is becoming more difficult. Senior engineers understand architectural intent, recognize subtle design flaws, anticipate long-term maintenance challenges, and identify security weaknesses that aren’t immediately visible. Those judgment skills become even more valuable when AI produces the first draft of nearly everything.

The same principle applies to quality assurance teams. Their role is evolving beyond finding bugs after development has finished. They’re becoming the guardians who determine whether AI-generated software actually satisfies the organization’s standards for security, reliability, performance, and maintainability.

Quality is no longer the final checkpoint before deployment. It’s becoming the discipline that allows organizations to embrace AI without sacrificing confidence in their software.

That’s why I believe every AI-generated pull request should be treated as a hypothesis rather than a finished product. The question isn’t whether the code runs. The question is whether it’s the right solution for the business, whether it strengthens or weakens the architecture, and whether it introduces risks that won’t become visible until months after deployment.

AI is transforming software development in remarkable ways, and I wouldn’t want to return to the slower world that existed before these tools arrived. Yet the lesson from Alexey Grigorev’s experience is impossible to ignore. AI can generate software at extraordinary speed, but it can also generate extraordinary mistakes. The faster we write code, the more disciplined we have to become about validating it.

LEARN MORE

Leave a comment

Leave a comment

Today, every company is under pressure to behave like a technology company, regardless of its industry. AI has lowered the cost of building software, shortened product development cycles, and made experimentation dramatically cheaper. Competitive advantage no longer comes from writing code faster. It comes from identifying opportunities before everyone else and turning ideas into measurable business value.

The Entrepreneurial Mindset

McKinsey’s Global Tech Agenda 2026, based on responses from more than 600 technology and business leaders, found that technology executives at the highest-performing organizations are deeply involved in business strategy, not just technology strategy. Nearly two-thirds of top-performing companies reported that their technology leaders play a significant role in shaping enterprise strategy. Those organizations also collaborate continuously between business and technology teams instead of relying on annual planning cycles.

Entrepreneurs begin with problems rather than technology. They ask where customers are frustrated, where markets are shifting, and where competitors have become complacent. Technology becomes the tool rather than the objective.

Great tech leaders are beginning to ask the same questions. Instead of asking, “How do we implement AI?” they’re asking, “What customer problem becomes easier to solve because AI exists?” That’s a completely different mindset.

Traditional technology leadership rewarded operational excellence. Modern technology leadership rewards opportunity recognition.

Leave a comment

Being Comfortable with Uncertainty

Entrepreneurs become comfortable with uncertainty because they know they won’t have perfect information before making decisions. They launch minimum viable products, collect feedback, adjust, and improve. Waiting for certainty usually means arriving too late.

AI is pushing enterprise technology in exactly that direction. Development cycles that once took months now take weeks. Some prototypes take days. Organizations that insist on exhaustive planning before releasing anything will increasingly lose to competitors willing to experiment intelligently.

McKinsey’s research on corporate venture building reinforces this point. Its 2025 global survey found that organizations that repeatedly build new businesses consistently outperform those treating innovation as an occasional event. Experience compounds. Teams become better at identifying opportunities, testing assumptions, and scaling successful ideas. AI accelerates that entire cycle by reducing both development costs and experimentation time.

How Entrepreneurs View Failure

Entrepreneurs see failures differently. A failed experiment that costs two weeks and prevents a $20 million mistake isn’t failure at all. It’s inexpensive market research. That’s becoming one of AI’s greatest strategic advantages. The cost of testing ideas has collapsed.

Instead of debating whether customers might want a feature, teams can build a prototype, place it in front of users, and learn from real behavior. That feedback becomes more valuable than internal opinions.

Another entrepreneurial characteristic is resourcefulness. Startup founders rarely have enough money, enough people, or enough time. They learn to create leverage by combining existing assets in unexpected ways.

Technology leaders face the same challenge. Every CIO wants a larger budget. Every CTO wants more engineers. Every executive wants additional AI investment.

The entrepreneurial technology leader asks a different question. “What can we accomplish with the resources we already have?” Generative AI has become the ultimate force multiplier.

Routine coding, documentation, testing, and analysis are increasingly automated. That frees technical talent to spend more time solving higher-value business problems. Microsoft’s newly introduced EngThrive framework reflects this shift by measuring engineering organizations across speed, quality, ease, and developer well-being rather than focusing only on output metrics. The emphasis is moving from activity to outcomes.

Leave a comment

An Obsession with the Customer

Many technology organizations still define success by completing projects. Entrepreneurs define success by changing customer behavior. Those measurements aren’t interchangeable.

Shipping software means nothing if adoption never follows.

The best technology leaders increasingly spend time with customers, sales teams, marketing leaders, and frontline employees. They search for friction that technology can remove rather than projects that technology can complete. That’s exactly how founders discover opportunities.

Perhaps the strongest evidence that technology leadership is becoming entrepreneurial comes from AI itself.

McKinsey recently argued that AI is turning every company into a software company. As software becomes easier and cheaper to create, the real constraint shifts from engineering capacity to business imagination. The winners won’t be the organizations that generate the most code. They’ll be the ones that identify the most valuable problems worth solving.

The future CTO, CIO, and Chief AI Officer won’t be judged primarily by uptime, delivery schedules, or infrastructure modernization. They’ll be judged by revenue created, markets entered, customer problems solved, and competitive advantages built.

Technology expertise remains essential. But business expertise is becoming mandatory. Entrepreneurship connects the two.

The organizations that thrive over the next decade won’t separate technology leaders from entrepreneurs. They’ll expect their technology leaders to become entrepreneurs.

LEARN MORE

Leave a comment

Leave a comment

I’ve watched companies roll out AI initiatives with all the right language and none of the real intent. Leadership gets excited about possibilities, middle management gets cautious about disruption, and the teams on the ground are left trying to fit something new into a system that was never designed to support it. A few pilot programs show promise, some efficiency gains appear, and then the entire effort gets absorbed back into the old way of operating. The structure wins every time.

That’s the part no one wants to admit. Most organizations don’t want to change. They want to improve just enough to stay competitive without disturbing the internal balance that keeps everything predictable. AI becomes a layer on top of the business instead of a force that reshapes it, and that’s why so many of these efforts stall out long before they deliver anything meaningful.

AI Doesn’t Transform—It Exposes

If you want to become a real change agent in the age of AI, you have to start by rejecting that entire approach. You can’t treat AI like an upgrade to an existing system because the system itself is the problem. The processes, the approvals, the incentives, all of it was built for a different era, and layering intelligence on top of it only makes the cracks more visible.

Most people aren’t comfortable with that level of honesty. They’d rather focus on tools because tools feel manageable. You can learn them, measure them, and present them in a way that looks like progress. Questioning the system is different. It creates tension, and tension is something most organizations are designed to avoid.

What AI really gives you isn’t just speed. It gives you clarity. It exposes redundancy in ways that are hard to ignore, and it highlights how much of the work inside an organization exists simply because it always has. When you start to see that, it changes how you think about everything from decision-making to ownership.

You Have to Choose What to Break

At that point, you have a choice to make. You can use AI to reinforce what already exists, or you can use it to challenge it. One path keeps you aligned with the organization, and the other puts you at odds with it. Only one of those paths leads to real change.

I’ve learned that the people who drive transformation aren’t the ones with the deepest technical knowledge. They’re the ones who are willing to question why things are done the way they are. They look at a workflow and ask whether it still makes sense, and they don’t accept “because we’ve always done it this way” as an answer.

That mindset creates friction, and friction is unavoidable if you’re serious about changing anything that matters. Organizations resist change because it threatens existing power structures, and AI has a way of accelerating that threat. It doesn’t just make work more efficient/ it reveals where control is being held for the wrong reasons.

Where Most AI Initiatives Quietly Die

That’s where most initiatives break down. The technology works, but the organization quietly rejects what the technology exposes. People protect their roles, their influence, and their version of how things should operate, even when it’s clear that a better approach exists.

If you’re stepping into this environment as a change agent, you can’t ignore that dynamic. You have to move directly into it. That means pushing beyond pilot programs and forcing successful ideas into the core of the business, where they actually have an impact. It means taking something that works in isolation and proving that it can outperform the existing system at scale.

That’s when resistance becomes visible.

Don’t Wait for Alignment—Create It

I’ve seen teams build AI-driven capabilities that could fundamentally improve how a company operates, only to have those capabilities sidelined because they didn’t fit neatly into existing approval structures. I’ve seen leaders celebrate innovation in public settings while quietly slowing it down behind the scenes because it disrupted their span of control. None of this is accidental, and none of it is new.

What is new is the speed at which these tensions are surfacing.

AI compresses time in a way that organizations aren’t used to. It lowers the cost of experimentation and increases the pace at which results show up, which means the gap between what’s possible and what’s actually happening becomes impossible to ignore. That gap is where change agents operate.

They don’t wait for consensus because consensus usually forms around maintaining the current state. Instead, they create undeniable proof that a different way works better. They build, test, and implement faster than the organization can debate, and they let results force the conversation.

AI Is Pressure, Not the Solution

The mistake most people make is assuming that AI is the change itself. It isn’t. AI is the pressure that forces change to happen. It exposes inefficiencies, accelerates feedback loops, and makes weak thinking more obvious. If you’re not willing to confront those things directly, AI won’t help you. It will simply highlight the problems you’re avoiding.

Becoming a change agent in this environment requires a different posture. You have to be willing to challenge assumptions that have been in place for years, and you have to do it in a way that produces tangible results. Talking about change isn’t enough. You have to demonstrate it, even when it creates discomfort.

There’s no version of this where everyone is aligned from the beginning. Some people will resist you because the changes you’re driving will alter how they operate or reduce the control they currently have. That resistance is a signal that you’re pushing on something real.

The companies that come out ahead in this era won’t be the ones with the most advanced AI models. They’ll be the ones willing to rethink how their business actually functions. That shift doesn’t happen through consensus or careful alignment. It happens because someone decides the current state isn’t good enough and starts building something better.

If you’re waiting for permission to do that, you’re already behind.

LEARN MORE

Leave a comment

Most organizations approach artificial intelligence the same way they approach any new technology. They look for cost savings, productivity improvements, and return on investment. Those metrics matter, but they only tell part of the story.

Leave a comment

Leaders are asking the wrong question about AI. They want to know how much value artificial intelligence can create for the organization. That question matters, but it’s incomplete. A better question is this: How much value can AI create, how effectively are we using it, and what are we risking in the process?

AI value can’t be measured by productivity gains alone. It can’t be reduced to saved hours, lower costs, faster workflows, or fewer employees. Those numbers may look good in a board presentation, but they don’t always tell the truth. A company can move faster and become sloppier. It can cut costs and lose institutional knowledge. It can automate processes and weaken the judgment of the people who once understood the work.

Benefit. Utilization. Consequences.

The first measure is value creation. This is where most companies begin, and for good reason. AI can help teams work faster, reduce repetitive tasks, improve analysis, draft content, summarize documents, write code, support customers, and identify patterns buried inside large amounts of data. These gains are real and we should measure them.

The mistake is stopping there. A company that uses AI to save ten hours a week hasn’t automatically created strategic value. The real question is what happens with those ten hours. Are employees using that time to deepen customer relationships, improve products, make better decisions, and pursue growth? Or are those hours disappearing into more meetings, more noise, and more low-value activity?

AI Utilization & Efficiency

Leaders shouldn’t only ask whether AI creates value. They should ask whether the organization is using AI efficiently enough to capture that value. That means looking at adoption, training, workflow integration, quality control, and use-case discipline. Are employees using AI in ways that improve outcomes, or are they just experimenting without direction? Are teams applying AI to meaningful business problems, or are they forcing it into places where it doesn’t belong? Are managers measuring the quality of AI-assisted work, or only the speed of completion?

AI utilization efficiency is the difference between owning a powerful tool and knowing how to use it. Many organizations will buy AI platforms, announce AI initiatives, and still fail to gain meaningful advantage because they never redesign the work around the technology. They add AI to broken processes and expect transformation. That doesn’t work. It only makes poor systems move faster.

Leaders should begin with clear use cases. AI should be tied to a real business problem, not vague pressure from the market. If a company adopts AI because competitors are doing it, the initiative starts from fear rather than strategy. That usually leads to scattered pilots, weak governance, and inflated expectations.

A solid use case answers basic questions. What problem are we solving? Who benefits? What processes change? What improves? How will we know? What could go wrong? Without those answers, AI becomes theater. It looks innovative from the outside, but inside the company, it creates confusion.

What Are the Consequences?

Every AI gain should be measured against its potential harm. Speed is a good example. AI helps organizations move faster, but leaders should ask whether they are moving too fast. Faster research can support better decisions, but it can also create false confidence. Faster content production can increase output, but it can also flood the market with generic thinking. Faster software development can accelerate innovation, but it can also introduce errors, security risks, and technical debt.

The same applies to cost reduction. AI can reduce headcount, but leaders should be careful about what else gets lost. When experienced people leave, they take judgment, context, relationships, and cultural memory with them. Those assets rarely appear on a spreadsheet, but companies feel their absence when things get difficult.

This is one of the hidden dangers of measuring AI only through efficiency. An organization may look leaner while becoming weaker. It may process work faster while losing the human knowledge that helped it interpret complexity. It may reduce payroll while damaging morale, trust, and collaboration.

Corporate culture also matters. If employees experience AI as a threat, adoption suffers. If they see it as a tool that helps them do better work, adoption improves. Leaders need to measure trust, not just usage. They should ask whether AI is strengthening people or making them feel replaceable. The answer will shape how deeply the technology becomes embedded in the organization.

The best AI measurement system balances four questions.

1. What value did AI create?

2. How efficiently did we use AI to capture that value?

3. What did we gain in speed, cost, quality, insight, or capability?

4. What did we risk, weaken, or lose along the way?

That last question may be the most important one. AI can make an organization more intelligent, but only if leaders use it with discipline. It can amplify judgment, but it can also replace judgment with automation. It can increase capacity, but it can also create dependency. It can help people do better work, but it can also flatten the human expertise that makes the company distinct.

The real measure of AI value is not whether the organization becomes faster. It is whether the organization becomes better.

LEARN MORE

Leave a comment

Share

The metric that matters isn’t how many lines of code AI can generate. It’s how much software gets successfully delivered, adopted by users, and translated into business outcomes. That distinction is becoming harder to ignore.

A recent MIT-led analysis found that AI tools dramatically increased coding activity. In some environments, file edits nearly tripled after AI coding assistants were introduced. Yet much of that increased activity disappeared as projects moved through testing, review, deployment, and adoption. The researchers found that while coding output surged, software releases increased by only about 30 percent. Even more striking, many of those additional releases failed to generate meaningful user engagement or business value.

In other words, AI is accelerating code production faster than it’s accelerating value creation.

That finding aligns with what our team at SONATAFY has seen across the industry. Many leadership teams are celebrating activity metrics because they’re easy to measure. They track commits, pull requests, story points, and development velocity. AI makes all of those numbers look better. A developer who once produced ten units of output can now produce twenty.

Code Generation is Not the Problem

The problem is that software development has never been a code-generation problem. It’s about problem-definition, prioritization, customer-adoption, and ultimately business-value. Code is only one step in a much larger chain.

Imagine a construction company measuring success by the number of bricks delivered to a job site. If the building is never completed, nobody would call that productive. Yet that’s effectively what many organizations are doing with AI-generated code. The danger is that AI creates an illusion of progress. When executives see development activity rising, they naturally assume more value is being created. The dashboards look impressive. The engineering organization appears more productive. Teams are shipping more artifacts into repositories.

Yet customers don’t buy code. Customers buy solutions. A feature that nobody uses has exactly the same business value whether it required ten lines of code or ten thousand.

That’s why some of the latest research on AI productivity is so important. Another study conducted by the research organization METR found that experienced developers working on familiar codebases actually took longer to complete tasks when using AI coding tools. Even more surprising, developers believed they were becoming more productive when the measured results showed the opposite.

That finding highlights a broader challenge. AI can make work feel faster because it reduces the effort required to generate output. However, software engineering involves much more than generation. Developers still need to review code, validate assumptions, test integrations, resolve edge cases, address security concerns, and maintain the resulting systems. As AI-generated code volumes increase, so does the burden of verification.

Organizations that focus solely on code output risk creating what I call synthetic productivity. Activity rises. Repositories grow. Dashboards improve. Yet the underlying business outcomes remain unchanged. The better question isn’t how much code AI generated. The better question is what value that code created.

The companies that benefit most from AI won’t be the ones generating the most code. They’ll be the ones that become disciplined about connecting software development to measurable outcomes. AI has given us the ability to manufacture code at industrial scale. That’s a remarkable achievement. The challenge now is ensuring we don’t confuse production with progress.

History is filled with examples of organizations that optimized the wrong metric. Factories produced inventory nobody wanted. Sales teams generated leads that never converted. Marketing departments celebrated clicks that never became customers. Software development is facing a similar moment.

Steve Taplin

Leave a comment

WANT TO LEARN MORE? DOWNLOAD YOUR FREE REPORT.

I recently read McKinsey’s latest CEO Playbook for Leading Through Uncertainty, and one idea stood out above the rest. The organizations that thrive over the next decade won’t be those that respond to disruption. They’ll be the ones that reinvent how they operate before disruption forces them to.

Leave a comment

No executive faces that challenge more directly than the Chief Technology Officer.

For years, the CTO playbook was well understood. Build a talented engineering organization. Adopt Agile. Embrace DevOps. Move to the cloud. Automate testing. Accelerate software delivery. Improve developer productivity.

Those practices transformed software development. Now, Artificial intelligence is changing the equation.

AI isn’t another development tool. It is a new participant in the software development lifecycle. It writes code. It recommends architectures. It creates documentation. It generates test cases. It identifies defects. Increasingly, it performs work that only human developers performed a few years ago.

The New Software Risks

AI is creating tremendous opportunities. It’s also creating an entirely new category of software risk.

Much of the current discussion around AI focuses on productivity. Research from GitHub has shown that developers using AI coding assistants complete many programming tasks significantly faster. Organizations everywhere are investing in AI to increase engineering velocity and reduce development costs.

Speed has never been the ultimate measure of software success. Quality is.

The 2025 DORA Accelerate State of DevOps Report delivered an important warning to technology leaders. While AI improved individual developer productivity, higher levels of AI adoption did not consistently improve organizational software delivery performance. In several cases, software stability, delivery performance, and developer satisfaction declined when AI was introduced without corresponding changes to engineering practices.

Artificial intelligence is allowing organizations to generate more code than ever before. At the same time, it introduces risks that traditional software development frameworks were never designed to address.

• AI can hallucinate nonexistent functions

• It can recommend insecure programming patterns

• It can generate inconsistent coding standards across teams

• It can introduce hidden technical debt that developers fail to recognize because they did not write the original code

• It can create intellectual property concerns when developers unknowingly incorporate generated content with uncertain licensing histories

• Most importantly, AI can produce software that appears correct while concealing flaws that become visible only after deployment

Traditional software quality frameworks were built for software created almost entirely by humans. Tomorrow’s software will be built by humans working alongside intelligent machines. That demands a different quality framework.

The New CTO Playbook

The new CTO playbook is no longer about maximizing coding productivity. It is about maximizing confidence in AI-assisted software delivery. That begins with a new AI Software Quality Framework built around six core principles.

1. Human Accountability

Every AI-generated artifact should have an accountable human owner. AI can generate code. Humans remain responsible for approving what enters production.

2. Architecture Integrity

Large language models excel at solving isolated programming problems. They don’t understand long-term enterprise architecture the way experienced software architects do. Every AI contribution should be evaluated within the broader system design rather than accepted in isolation.

3. Continuous Security Validation

Every AI-generated component should undergo automated security scanning, dependency analysis, vulnerability testing, and compliance review before deployment. Security can no longer be a checkpoint at the end of development. It becomes a continuous discipline integrated throughout the software lifecycle.

4. AI Verification

Testing now extends beyond verifying functionality. Teams must validate AI-generated business logic, identify hallucinated code, evaluate model-generated recommendations, and confirm that software behaves consistently across changing AI outputs.

5. Governance and Traceability

Organizations need to know where AI contributed to software development, which models generated the code, who approved it, and how decisions were validated. As regulations evolve, that visibility will become a strategic requirement rather than an operational convenience.

6. Continuous Organizational Learning

Every production defect, security incident, and architectural failure should improve both the engineering organization and its AI workflows. AI systems learn from feedback. Engineering organizations should do the same.

Collectively, these principles redefine the CTO’s role. Tomorrow’s technology leaders will spend less time asking, “How can AI help us write more code?” Instead, they will spend more time asking, “How do we ensure every line of AI-generated code meets the quality, security, reliability, and governance standards our customers expect?”

That’s a different leadership challenge. The next decade won’t belong to the organizations generating the most software. It will belong to the organizations delivering the most trusted software.

The last twenty years gave us Agile, DevOps, and continuous delivery. Those frameworks transformed how humans build software. The next era requires something equally important: a comprehensive AI Software Quality Framework that governs how humans and artificial intelligence build software together.

The CTOs who embrace that challenge today will do more than deploy AI successfully. They will define the new standard for software excellence in the age of artificial intelligence.

Leave a comment

WANT TO LEARN MORE? DOWNLOAD OUR FREE REPORT

Leave a comment