The Most Rational Take on AI: What SMB Leaders Should Actually Do Next

Artificial intelligence is generating two kinds of reactions at the same time. One is excitement so intense it borders on prophecy. The other is anxiety so intense it slides into paralysis. Between those poles, many business leaders, especially in small and midsize businesses, are left asking a more practical question: what should we actually do now?

That is why a grounded, rational perspective matters so much. In a recent conversation, tech analyst Benedict Evans offered one of the clearest frames business leaders could ask for. His core point was both bold and calming: AI is as big a deal as the internet or mobile, and only as big a deal as the internet or mobile. That may sound like a hedge, but it is actually a powerful strategic lens.

“AI is as big a deal as the internet or mobile and only as big a deal as the internet or mobile.”

For iAvva AI Consulting, this is exactly the kind of framing organizations need right now. AI is a profound shift. It will change workflows, jobs, expectations, software, and competitive dynamics. But it is not helpful to treat every new model release as the end of work or the beginning of machine destiny. Businesses need a more useful question: where will AI create real value, and how should leaders prepare their people and systems to use it well?

Why This Framing Is So Useful

Comparing AI to the internet or mobile is not minimizing it. Those were civilization-scale technology shifts. They changed communication, commerce, media, work, logistics, and personal behavior. Saying AI is in that class is already saying it is enormous.

What Evans pushes back on is the temptation to exaggerate beyond usefulness. Business leaders do not need an endless debate about whether AI is 20% bigger or 200% bigger than previous platform shifts. They need to know what kind of planning mindset to adopt. The right mindset is this: the shift is real, the path is uncertain, and the practical winners will be the organizations that experiment intelligently without panicking.

That is especially true for SMB leaders in IT, HR, operations, and transformation roles. Most are not trying to build AGI. They are trying to improve workflows, reduce friction, support employees, and create measurable business outcomes.

We Are Early, Even If It Does Not Feel Like It

One of the most important ideas in the conversation is the analogy that AI feels like the internet in 1997. The technology is exciting. The pace feels dizzying. Some tools already feel transformative. But most of the things people will ultimately do with AI have not been built yet. And many of the current products are still uneven, brittle, or unclear in their business role.

That matters because it explains why so many leaders feel both urgency and confusion. The market is noisy because we are early. There is a difference between a foundational shift being real and the deployment pattern being mature. Those are not the same thing.

Why “The Job Apocalypse” Is the Wrong First Question

One of the strongest parts of the discussion is the rejection of simplistic labor panic. Evans’ point is not that disruption is painless. It is that history shows a more complicated pattern. New technologies automate certain tasks, eliminate some roles, create frictional pain, and then unlock new categories of work that did not exist before.

This does not mean leaders should be casual about displacement. It does mean they should be skeptical of sweeping claims that every company will buy one AI system and fire half its staff two weeks later. That is not how organizations actually work. Enterprise systems, workflows, risk policies, and human adoption all slow change down, even when the technology itself moves quickly.

For business leaders, the lesson is not “do not worry.” The lesson is “worry more intelligently.” Focus less on abstract apocalypse and more on concrete exposure:

• Which tasks are repetitive and documentation-heavy?
• Where are workflows slow or fragmented?
• Which support questions repeat constantly?
• What decisions require better access to information?
• Where could employees benefit from guided assistance rather than full replacement?

That is where implementation starts to become useful.

The Real Opportunity Is in the Workflow Layer

One of the most important practical insights in the transcript is the idea that the model alone is rarely the full solution. The real work of transformation happens in what surrounds the model: the harnesses, the applications, the integrations, the workflow redesign, the human support, and the context.

That is exactly where iAvva AI Consulting can create value. Most organizations will not win by obsessing over which frontier lab is philosophically closest to AGI. They will win by asking better implementation questions.

For example:

• How should AI assist onboarding without weakening trust?
• How can a knowledge assistant reduce repetitive HR and IT burden?
• Which reporting workflows should become more intelligent first?
• How do we support change adoption so the tool becomes habit, not novelty?
• How do we tie AI use to measurable outcomes rather than vague experimentation?

These are the kinds of questions that move AI from fascination to business value.

What Business Leaders Are Still Not Pricing In

A rational take on AI does not just reject hype. It also highlights what many organizations are underestimating.

First, leaders may be underestimating how much software will change. Evans points out that software development already appears to be in a before-and-after moment. Even if other professions are earlier in their adoption curve, coding, prototyping, and product development are already feeling the shift. That means the software stack around the enterprise will evolve quickly, even if the enterprise itself moves slower.

Second, leaders may be underestimating how much implementation labor is still required. Ironically, the rise of AI is not eliminating the need for consultants, forward-deployed engineers, and transformation support. In many cases it is increasing that need. Why? Because organizations do not have idle teams waiting to redesign internal workflows, connect systems, define governance, and retrain staff. Someone still has to do that work.

Third, leaders may be underestimating the gap between task automation and job transformation. A profession is rarely reducible to a neat percentage that can be automated. The real question is what part of the job is the task, and what part is judgment, trust, context, politics, or customer understanding.

Comparison: Panic Response vs Strategic Response

So What Should SMB Leaders Actually Do?

This is where the transcript becomes especially useful. The recommendation is not to hide from AI and not to romanticize resistance. Leaders should dive in, learn what today’s systems can do, and come out with a sharper understanding of where those systems help and where they still fail.

For SMB leaders, that can translate into a practical five-step agenda:

1. Build literacy at the leadership level. If decision-makers do not understand the current reality of AI, they will either overreact or underreact.
2. Choose one or two business workflows to improve. Avoid abstract company-wide mandates. Start with a support, reporting, onboarding, or knowledge workflow.
3. Define the human role clearly. Decide what stays with people, what AI can assist, and where escalation matters.
4. Measure outcomes early. Track time saved, support burden reduced, quality improved, or adoption increased.
5. Treat adoption as a people challenge, not just a software launch. This is especially important in HR, IT, and transformation work.

That is a much stronger response than either panic or passivity.

Why This Matters for HR and IT Leaders

HR and IT leaders sit at a crucial intersection in this transition. IT understands systems, data, security, and technical feasibility. HR understands communication, behavior, trust, training, and organizational readiness. AI will succeed much more often when those two functions work together rather than separately.

If AI is introduced only as infrastructure, adoption will suffer. If it is introduced only as aspiration, reliability will suffer. The winning model is integrated: build useful systems and help people use them well.

The iAvva AI Consulting Perspective

At iAvva AI Consulting, we believe the most rational path is also the most strategic one. AI should be treated neither as magic nor as menace. It should be treated as a powerful new layer of capability that must be translated into workflow, behavior, and measurable business results.

That means helping leaders answer questions like:

• Which use cases are practical now?
• Where should custom AI solutions be used instead of generic tools?
• How do we improve adoption across teams?
• How do we implement AI responsibly without freezing progress?
• How do we move from curiosity to capability?

These are not side questions. They are the real work of transformation.

Key Takeaways

• AI is a major platform shift, but not every part of the economy will change at the same speed.
• Panic about instant job collapse is less useful than careful analysis of task exposure and workflow redesign.
• The biggest value will likely come not just from models, but from applications, integrations, and implementation.
• HR and IT leaders play a central role in making AI both useful and trusted.
• The strongest business response is deliberate experimentation combined with real adoption support.

Final Thought

The most rational take on AI is not boring. It is liberating. It gives leaders permission to stop performing certainty and start building capability. You do not need to predict the entire future of AGI to make smart decisions this quarter. You do need to learn, experiment, choose wisely, and help your people adapt.

That is how organizations move from AI anxiety to AI advantage.

Call to Action: If your organization is trying to cut through AI hype and focus on practical implementation, iAvva AI Consulting can help you identify the right use cases, design custom AI solutions, and support the human adoption that makes transformation stick.