Will AI 10x Your Productivity? An Honest Look at the Claim
By Stacey Tallitsch | September 21, 2026
You have heard the number. AI is going to make you 10x more productive. Do the work of 10 people. Get your nights and weekends back. The claim shows up in ads, in webinars, and in the mouth of every consultant trying to sell you something.
It is a big promise. And you are busy enough that a promise like that is hard to ignore. If a tool could really do 10 times the work, you would want to know why you are not using it yet.
So let me give you the honest version. Not the sales version, and not the doom version. Just what the research actually shows about how much time AI saves, and where that "10x" number falls apart.
Where the "10x productivity" claim comes from
The claim is not made up out of nothing. There is something real underneath it, and it is worth naming before we take it apart.
When a task is a good fit for AI, the speed-up is real and worth having. AI is good at first drafts. It is good at reading a long document and pulling out the main points. It is good at answering the same routine question 100 times without getting tired. For that kind of work, a 30-minute task can turn into a 10-minute task.
Picture a contractor who hates writing follow-up emails to customers who have not paid. The first time an AI tool drafts one of those emails in 20 seconds, it feels like magic. Or a salon owner who spends her Sunday night replying to online reviews one by one. The first time a tool drafts all of them at once, she gets an hour of her weekend back.
That first-time feeling is powerful. And the people selling AI take that feeling and stretch it into a promise about your whole business. One task got 3 times faster, so surely your whole operation can get 10 times faster. That is the leap. And that is where it breaks.
What the research actually found
Real studies have measured this, and the numbers are a lot smaller than 10x.
Economists at Stanford, MIT, and the National Bureau of Economic Research ran one of the largest studies on the subject. They watched more than 5,000 customer support agents use an AI assistant on real calls, over many months. The AI helped. Productivity went up by about 14%. The study, called Generative AI at Work, is the kind of primary source worth trusting. It measured real work over many months, not a demo built to impress you.
14% is a good result. It is not nothing. But 14% is not 1,000%. The distance between what the ads promise and what the research measured is enormous.
There is a second finding in that same study that matters even more for you. The biggest gains went to the newest and least experienced workers. The most experienced agents barely gained at all. AI raised the floor. It did not raise the ceiling. If you are already good at the core work of your business, AI helps you less. It does more for a brand new hire who is still learning the ropes.
The part nobody selling AI wants to admit
Here is the finding that should stop you cold.
A research group called METR ran a careful test with experienced software developers in 2025. These are people who use AI tools every day and know them well. The researchers gave them real work to do, some with AI and some without, and measured the actual time on the clock.
The developers were 19% slower when they used the AI tools. Slower, not faster. And here is the twist that should stick with you. After the test, those same developers guessed that the AI had made them about 20% faster. They were wrong about their own work by nearly 40 percentage points. You can read the write-up of that developer productivity study if you want the full details.
Stop and think about what that means for a second. Smart, skilled people felt faster while they were actually slower. AI feels productive. The typing is fast. The answers come back in seconds. There is always something on the screen to react to. But feeling productive and being productive are not the same thing, and most people cannot tell the two apart from the inside.
What is true, and what is oversold
Let me line this up plainly so you can plan around it.
What is true: AI can save real time on specific, repeatable tasks. Drafting emails, summarizing long threads, answering routine questions, cleaning up rough notes into something readable. In the right spot, you will feel the difference inside a week.
What is oversold: the idea that this adds up to 10x for your whole business. It does not. The real gains land somewhere between "a little slower" and "about 40% faster," depending on the task and the person doing it. The honest average across careful studies is a modest boost, not a miracle.
What is misleading: the promise that AI replaces the skilled part of your work. The research points the other way. AI helps most with the routine work. It helps least with the hard judgment calls that actually run your business. Think pricing a tricky job, reading a difficult customer, or deciding which work to turn down. It is a helper for your junior tasks, not a replacement for your senior thinking.
This is also why most small businesses do not need anything fancy or custom built to get the gains that are actually on the table. If you are wondering whether your shop needs to build its own system, the short answer is almost always no. I went through that in more detail in whether your business needs to build its own AI.
What to actually do about it
You do not need to chase 10x. You need to find the two or three tasks in your week that AI is genuinely good at, and hand those over.
Start with the boring, repeatable writing. Follow-up emails. First drafts of quotes or proposals. Summaries of a long email chain you do not want to reread. How you ask the tool matters more than most people expect, which I covered in what a prompt is and why the way you ask matters. Pick one task, use AI on it for a week, and time yourself honestly.
That last part is the whole game. Do not trust the feeling. Trust the clock. Write down how long the task took you before, and how long it takes with the tool. If the clock says you saved time, keep the tool for that task. If it says you did not, drop it and move on to the next one.
That is the entire method. It is not exciting. It will not fit on a webinar slide. But it is real, and it will beat any consultant promising to 10x your whole business by next quarter.
And if a competitor's big AI announcement is what has you worried in the first place, that pressure is usually worth less than it sounds. I wrote about why in why a competitor's AI announcement matters less than you think.
The honest takeaway is short. AI is a useful tool that saves real time on the right tasks. It is a bad bet for anyone hoping it will do 10 times the work of your whole team. Use it for what it is, measure it with a clock instead of a feeling, and ignore the number.
-- Stacey | The Standalone
About the Author
Stacey Tallitsch runs The Standalone, an AI Implementation Diagnostic practice for small business owners. He has 30 years of experience in technology and has written 21 books on systems thinking and decision-making. More than 30,000 students have learned from his online courses.
- Stacey Tallitsch, The Standalone