The AI Problem

AI is not struggling because people cannot see the threat.

It is struggling because the threat is the clearest part of the story.

When people talk about AI and work, the conversation gets pulled toward replacement almost immediately. Fewer people. More software. Less headcount. More automation. Higher margins. Better output per person.

Some of that is uncomfortable because some of it is true.

You cannot have a serious conversation about AI inside companies and pretend labour does not sit in the middle of it. If a team can do more with fewer manual steps, fewer handoffs, fewer meetings, fewer external vendors, or fewer roles built around moving information from one place to another, leadership will pay attention.

That does not make the message good.

It makes the message incomplete.

The bleak version travels faster

The easy AI story is the cold one.

“Companies are firing people, then using AI to replace the work.”

That story travels because people can already feel it. They have seen layoffs. They have watched job descriptions change. They have heard executives talk about efficiency in that polished way that usually means someone else is about to absorb more work.

So when the AI industry says “productivity,” a lot of people hear “job cuts.”

When it says “automation,” they hear “you are the manual step.”

When it says “agents,” they hear “a cheaper version of me.”

That is the narrative problem.

The economic reality is being explained without enough imagination.

We are asking people to accept a huge shift in how work gets done, but the loudest explanation often sounds like a layoff memo with better software attached.

No wonder people are uneasy.

The market understands the ratio

I do not think the market is only rewarding AI companies, cloud providers, chip makers, and large tech platforms because of hype.

Some of it is simpler.

AI points at a different ratio between people, software, and output.

A company can take expensive, slow, repeated knowledge work and compress parts of it into software. Support teams can answer more questions with fewer escalations. Product teams can prototype faster. Analysts can produce first drafts in minutes instead of days. Operations teams can stop treating every workflow as a bespoke manual process.

That does not mean every job disappears.

It does mean the old ratio changes.

More output per person.

More software in the middle.

More pressure on work that was only protected because the system around it was slow.

Companies will chase that. They already are. The question is whether AI can be sold as more than a headcount reduction machine.

Jobs sold possibility

The iPhone is the obvious comparison because it changed so many categories at once.

But the important part is not that Steve Jobs had better stage presence than everyone else.

The important part is that he sold possibility before disruption.

He did not walk on stage in 2007 and say, “This is going to hurt Nokia, BlackBerry, camera makers, GPS companies, MP3 players, mobile carriers, and half the software industry.”

That was all sitting inside the product.

The disruption was real. Whole categories were damaged or absorbed. Habits changed. Business models broke. Jobs changed.

But the story was not framed around loss.

The story was framed around what people could now do.

The internet in your pocket.

Music, phone, and web in one object.

Touch as the interface.

Software as the thing that made the device keep becoming new.

That framing mattered. It gave people a way to imagine themselves inside the change.

AI has mostly failed at that.

It keeps asking people to admire the machine before they can picture the better version of their own work.

Replacement is a weak public story

Replacement can be a business case.

It is a terrible public story.

Nobody wants to be invited into a future where the main promise is that fewer of them are needed.

I work in AI operations. I build with these tools. I think about adoption, governance, workflows, internal agents, and how non-technical teams can actually use this stuff.

I am not anti-AI.

But I do think the industry keeps making the least human version of the argument.

It talks about models, benchmarks, agents, token costs, speed, and automation. All useful. All part of the work.

Then it gestures vaguely at “the future of work” and expects people to fill in the upside themselves.

Most people will not.

They will fill in the downside because the downside is already visible.

The better story is capability

The strongest AI story is not:

“The company can do the same work with fewer people.”

It is:

“The cost of trying things is dropping.”

That changes the emotional shape of the technology.

A small team can build something that used to need a specialist function.

A non-technical operator can prototype a workflow instead of waiting for a backlog slot.

A founder can test more ideas before spending money they do not have.

A support team can see patterns in thousands of tickets without manually reading every one.

A creator can turn rough notes into a draft without pretending the blank page is a noble struggle.

An analyst can spend less time formatting the first pass and more time asking whether the answer is any good.

That is the upside.

Not fake optimism. Not “AI will free everyone to do creative work” as a blanket statement. That is too clean, and people do not trust it because they should not.

The better claim is more grounded:

AI reduces the cost of intelligence inside a system. When that happens, more people can attempt work that used to be blocked by time, skill, budget, or access.

That is a story people can step into.

The honesty has to stay

The trap is swinging too far the other way.

If AI companies copy the Jobs-style optimism while hiding the labour impact, people will smell it immediately.

The public does not need another glossy future pitch. It needs a more honest one.

Yes, some work will be automated.

Yes, some roles will shrink.

Yes, some companies will use AI badly and call it progress.

Yes, some leaders will chase cost reduction before they understand the system they are changing.

Say that plainly.

Then talk about the upside with the same level of clarity.

Because the upside is already showing up in the work.

I see it when someone who does not consider themselves technical builds a tool that actually helps their team.

I see it when a messy manual process becomes visible enough to improve.

I see it when AI stops being a chatbot and starts becoming part of the operating system around the work.

That is a better story than replacement.

It is also a more accurate one.

AI needs a better invitation

Messaging is not decoration.

It changes who feels invited into the shift.

If the story is only about replacement, the natural response is fear, resistance, or quiet disengagement. People protect their corner. They hide process knowledge. They avoid the tool because using it feels like training the thing that will be used against them.

If the story is about capability, the response changes.

People can ask better questions.

What parts of my work should not be manual anymore?

What could I build if the first version only took an afternoon?

What knowledge is trapped in my team because the current system makes it too hard to use?

What would I try if the cost of trying was lower?

That does not remove the hard parts.

It gives people a reason to participate before the hard parts are decided for them.

I do not think AI needs more hype.

It needs a better invitation.

The technology is already serious. The economics are already serious. The workforce impact is already serious.

But the story is still underbuilt.

Right now, too much of the public narrative sounds like this:

“This will replace a lot of work, but trust us, it will be good eventually.”

That is not good enough.

The better story has to hold two things at once.

AI will remove work.

AI will also make new work possible.

The second part needs more imagination than the first.

That is where the industry keeps falling short.

Because if AI is going to change the world, the story cannot just be that companies found a cheaper way to run the old one.

The story has to show what people can build next.

- JC