AI

More and more people use AI, but skills and company rules are not keeping up

New studies show that most small businesses and more than half of American adults already use AI, and in Poland almost every young adult does. At the same time, only a few people use it well, and companies often do not know which tools their staff rely on.

Bartosz Bielawski

Bartosz Bielawski

· 3 min read
Share
More and more people use AI, but skills and company rules are not keeping up

Signal in the data

Independent sources on this topic: 1 in the last week, 7 in 30 days, 15 in 90 days. We have tracked it since 2026-08-21; 26 claims in our database in total.

wk 320
wk 334
wk 340
wk 350
wk 361
wk 373
wk 383
wk 391
wk 400
wk 411
wk 424
wk 431

Number of independent sources per week (last 12 weeks).

What is happening

Over the past two weeks, a series of studies has come out on who uses AI and how often. The picture is consistent: AI use keeps growing, even among people who are worried about it.

In the US, 55% of adults use AI tools at least once a week. At the same time, 89% have at least one concern about AI (Search Engine Journal).

In Poland, only 36% of people used AI in the past three months. Among 18- to 24-year-olds, however, as many as 81% use it regularly (CallPage).

Where sources agree

Sources agree that AI is spreading through companies faster than the skills needed to use it. Two-thirds of small businesses in the US use AI regularly, and 28% use it daily (Logically Answered).

Most of that use goes to writing and marketing - 83% of small businesses that use AI use it for these tasks (Logically Answered).

On the other hand, estimates suggest that only about 5% of people in large companies can use AI to do genuinely useful work (Nate B Jones).

Where they differ

The biggest gap is not between sources but between people. Among weekly AI users, 52% say its impact is positive. Among those who never use it, only 6% say it is positive and 75% say it is negative (Search Engine Journal).

The sources also put the emphasis in different places. Some focus on the benefits and the pace of adoption, while others warn that the risk of AI security vulnerabilities is rising - 80% of World Economic Forum respondents say so (SekurakTV).

Earlier, sources described shadow AI: employees using AI tools without the company's knowledge and outside its rules (Michał Barczak).

What to do about it

Assume your people already use AI, even if nobody officially rolled it out. Instead of banning it, decide which tools are allowed and which data must never be pasted into them.

Invest in skills, not just licences. The gap between someone who "uses AI" and someone who actually does better work because of it is huge.

If you create content, remember that almost everyone uses AI to write. What sets you apart is your own data, your experience and examples that a model cannot make up.

Our take

AI has become an everyday tool for a large share of people and companies, including those who do not trust it. The question is no longer whether people use it, but whether they use it well and safely. The winners will be the companies that stop pretending AI does not concern them, teach their people to use it and set clear rules.

Confidence: medium - several independent studies point in the same direction, but they come from different countries, measure different things and are hard to compare directly. We will change our view if: new studies show that regular AI use in companies has stopped growing, or that the share of people who use it well has clearly caught up with the share who use it at all.

Sources

Read more