AI disclosure: this content, including any imagery, was artificially generated.
TL;DR: AI use is spreading, but unevenly. In a 2026 survey of 1,256 participants in Goldman Sachs’ US 10,000 Small Businesses programme, 76% said they use AI and 93% of those users said it had a positive impact. In the UK, the ONS puts use at around 35% of businesses with 10 or more employees and 28% of those with fewer than 10. Using AI and building it in are different things: in the same US survey, only 14% said AI was fully embedded in their core operations. None of the figures quoted here compare the earnings of owners with AI systems and those without, so ask how any quoted earnings gap was measured. A practical way forward is to pick one leak in your business, automate it properly and measure the result.
What the survey data actually shows
Two recent sources cover different populations, so read them side by side rather than as one picture.
In the US, a survey for Goldman Sachs, run by Babson College and David Binder Research between 27 January and 4 February 2026, collected responses from 1,256 participants in its 10,000 Small Businesses programme. That is a survey of one programme’s participants, not of all US small businesses. 76% said they currently use AI, and 93% of those using AI said it had a positive impact on their business. 84% cited increased efficiency and productivity as the primary benefit. Only 14% said AI was fully embedded in their core operations, and 73% said they would benefit from more access to training and implementation resources.
In the UK, the Office for National Statistics reports that around 35% of businesses with 10 or more employees used at least one AI technology in June 2026, up from around 12% in late 2023. Use rises with size: 28% of businesses with fewer than 10 employees, against 49% of those with 250 or more. Improving business operations was the most common use, at close to 60%.
What the data does not show
None of the figures quoted here compare the earnings of AI users with non-users. The Goldman Sachs figure is owners’ own view of AI’s impact, among owners who already use it. The ONS figures measure who uses AI, not what it earns them. Treat any precise “AI earnings gap” claim with caution unless it shows how it was measured.
Where owners get stuck
In the UK data, use rises with firm size. Beyond that, owners get stuck in three places:
- Depth. In the US survey, 76% said they use AI, but only 14% said it was fully embedded in their core operations.
- Skills. In the US survey, 73% said they would benefit from more access to training and implementation resources, and earlier ONS analysis found lack of expertise among the most cited UK barriers.
- A clear first use. Earlier ONS analysis also found difficulty identifying a business use for AI among the most cited barriers. A practical way forward is to start with one specific job.
Where to aim AI first
Aim AI at the leaks nearest the money first. Three are worth checking:
- Lead response: answering an enquiry in minutes rather than the next day.
- Follow-up: the quote chased and the proposal nudged by email or message, on a schedule that does not depend on anyone remembering.
- Admin: the hours lost to data entry and scheduling, which can cap how much the business can take on.
The mistake to avoid is waiting for one big, all-in-one AI project while the small, recurring leaks stay unfixed. If most of your enquiries arrive by phone, see our AI voice agents service for inbound calls.
A step-by-step roadmap
- Baseline: track where your time goes for a week, and note response times and follow-up rates. Set a target for each.
- Pick one leak: choose a repetitive, rule-based task close to revenue.
- Pilot it: automate that one task with a tool that fits your existing systems, keep a person checking the output, and check the data protection position (and marketing consent, if it contacts customers) before it touches client data.
- Measure: compare the same numbers before and after, over the same length of time.
- Embed: once it meets the targets you set, write it into how the team works, train the people who use it, then move to the next leak.
To set your baseline, use our time-tracking guide. To work out whether a task is worth automating, use the business automation ROI framework. For the specific cost-versus-benefit breakdown, see our AI automation cost versus savings analysis.
Pros and cons of adopting AI
| Possible gains | Costs and risks |
|---|---|
| Routine work handled faster and more consistently | Set-up time and cost |
| Staff time freed for higher-value work | Data protection and compliance duties |
| Faster responses to customers | Training needed before it sticks |
| Clearer data on how the business runs | Errors can spread if nobody checks the output |
| Fewer follow-ups forgotten or missed | Ongoing fees, maintenance and dependence on one supplier |
FAQs: Business Owners With AI Systems
Q: Do business owners with AI systems earn more than those without?
A: None of the figures quoted here measure that. Among US owners using AI in the Goldman Sachs programme survey, 93% said it had a positive impact, but that is their own view, not an earnings comparison.
Q: How many UK businesses use AI?
A: Around 35% of UK businesses with 10 or more employees reported using at least one AI technology in June 2026, according to the ONS, and 28% of those with fewer than 10 employees.
Q: What should I automate first?
A: One repetitive, rule-based task close to revenue, such as replying to new enquiries, chasing quotes or booking appointments. Measure it before and after, and keep a person checking work that needs professional judgement.
Q: What gets in the way?
A: Earlier ONS analysis found the most cited barriers were identifying a use, cost and lack of expertise. In the latest ONS data, 41% of UK businesses with 10 or more employees report no barriers, and cost affects about 7–14%. In the US survey, 73% said they would benefit from more access to training and implementation resources.
Related reading
- How dental practices are using AI to win more private patients
- The tasks eating the most time in your business and the AI fix for each
- Read more about how automation saves small businesses time
- AI marketing automation
Sources: Goldman Sachs 10,000 Small Businesses survey (press release, 17 March 2026); Office for National Statistics, “Artificial intelligence in UK businesses: 2023 to 2026” (20 July 2026). Last checked: 6 November 2026.
