Artificial intelligence has rapidly transitioned from the margins of business to the mainstream. More than half of UK firms (54 per cent) are now actively using AI in their operations, according to the British Chambers of Commerce – a vast leap from just 23 per cent three years ago.

“Consumers are using it non-stop, large corporates are heavily investing, and now SMEs too need to apply the widespread benefits to their businesses – to avoid falling behind,” says Ruchir Rodrigues, Client Strategy & Commercial Director at Lloyds Business and Commercial.

Two-thirds of UK businesses have invested in AI, according to the Lloyds Business Barometer, with most companies spending less than £25,000 on the technology. For many, these relatively modest investments are already translating into tangible day-to-day productivity gains.

“But our findings suggest only 7.5 per cent of SMEs engaging with AI are very confident in understanding the benefits for their business,” Rodrigues adds. “A lot of our one million business clients say, ‘it'll help productivity, but how do I put a number on it?’”

A quarter of small and medium-sized enterprises cite uncertainty around return on investment (ROI) as a primary barrier to AI adoption, according to a YouGov survey of over 1,000 IT decision-makers. However, Lloyds’ own customer research reveals that trust has emerged as an equally formidable hurdle.

In a crowded market filled with rapidly evolving players – including Google’s Gemini, OpenAI’s ChatGPT, Anthropic’s Claude, and Perplexity – business owners are struggling to know which tools to trust and where to safely invest.

This trust deficit is heavily compounded by deep concerns over data privacy and security, alongside a growing realisation that many firms lack the strong underlying data foundations needed to deploy these tools safely.

These issues stand alongside a persistent digital skills gap as critical barriers. For many SMEs, the hesitation isn't just about whether the technology works, but whether they can trust it with their proprietary data.

From soil mapping to automated fabrication

However, Rodrigues sees a more optimistic picture, thanks to the real-world use cases Lloyds’ 700 relationship managers across the UK are seeing British companies deploy.

“We have agricultural businesses using the technology to upload images of soil, analyse its content, and cross-reference weather patterns to decide which crops to plant each season – resulting in a more profitable harvest, year after year,” he explains.

Lloyds has partnered with Soil Association Exchange (SAX) to build a major farm environmental dataset, applying AI-driven mapping and analytics to generate insights on risk, resilience and investment opportunities.

In a second example, an advanced manufacturing client is building an entirely new business model around the technology.

“Clients upload images of what they want made, the AI generates a design and a 3D printer produces it,” Rodrigues adds.

The company has partnered with the Lloyds-backed Manufacturing Technology Centre (MTC) in Coventry, which offers expert support to help SMEs adopt advanced digital technologies such as AI-powered design tools, robotics and automation.

As Rodrigues points out: “Access to data and vastly improving processes are helping manufacturers come up with better solutions for swathes of industries.”

Navigating the AI maturity spectrum

For SMEs, AI maturity exists on a spectrum. At one end, businesses are simply dipping a toe in. Legal and professional services companies are using embedded AI systems to rapidly redact personal information from documents, for example, while others deploy large language models for content writing or to generate marketing ideas.

Indeed, more than six in 10 businesses are using AI for the creation of marketing content, according to a Lloyds customer survey*.

“In the middle, we see SMEs starting to integrate AI into processes to automate repetitive tasks,” Rodrigues adds, “while at the far end, the most sophisticated operators are building agentic AI systems: technology that doesn't just assist humans but independently executes complex tasks.”

Lloyds uses its own experience to help its customers with the technology. Its senior leaders are currently enrolled on a six-month programme with Cambridge University focused on agentic AI, helping the bank rank on the Global Evidence AI Index as one of the world's most AI-literate financial institutions.

More widely, Rodrigues represents Lloyds on the SME Digital Adoption Taskforce, a joint government-industry group specifically designed to help businesses break down these very barriers. By identifying practical, scalable solutions to overcome hurdles like the skills gap, security anxieties and ROI uncertainty, the taskforce acts as a vital bridge – helping SMEs confidently adopt productivity-enhancing digital technologies and accelerate their AI journey.

Internally, the bank saw the technology’s practical power when dealing with its thousands of property developer clients. Previously, these clients submitted tenancy statements and cash flow forecasts to secure lending in hundreds of different formats, meaning relationship managers devoted hours to reconciling the information.

“Now that’s all automated,” Rodrigues says. “It’s gone from a whole afternoon of work to a few minutes – across our whole business, that has a huge impact – and we’re helping clients roll out that productivity boost.”

Far from replacing human roles, this shift seems to have freed up relationship managers from administrative bottlenecks, allowing them to dedicate more time to strategic, face-to-face client support.

Why fundamental data is key

There is, however, a sobering note to Rodrigues' optimism: AI is only as good as the data feeding it.

“People get caught in the hype of AI but forget that fundamental data and strong governance is key: without that, it falls on its face.”

For AI to scale safely and sustainably, businesses must move beyond experimentation and implement rigorous controls, safeguards, and responsible use policies. Establishing these frameworks is what ultimately builds long-term customer trust.

Lloyds uses its own scale to lead by example. The bank is currently undertaking a significant cloud migration to unlock the value of its data. Crucially, the bank does not feed customer data into public models; instead, it adopts models from the likes of OpenAI or Google and pulls them inside its own secure systems, running the models internally to guarantee data privacy.

For an SME, leading by example looks remarkably similar, just on a different scale. Instead of massive custom infrastructure, it means ensuring that internal documents, client emails and financial spreadsheets are cleanly organised within secure, commercial versions of everyday software. By opting for business-grade subscriptions, a small firm can ring-fence its data just as securely as a major bank, ensuring proprietary information never leaks into the public domain.

Bridging the trust and capability gap

For smaller businesses without that scale of infrastructure, navigating these governance and security hurdles can feel overwhelming. Lloyds’ approach is sequential: build the data strategy first, then find models you can use internally. But the bank isn’t leaving SMEs to figure it out alone. Instead, it is actively deploying its network, finance, and expertise to bridge the trust and capability gaps.

“You want everyone in your business to know about AI, but you also need a small pocket of expertise that understands it deeper: get an expert in your business,” says Rodrigues.

To help businesses build that internal capability, the Lloyds Bank Academy provides free, on-demand online skills training and masterclasses. This practical guidance cuts through the market noise, showing SMEs how to build robust AI strategies while explicitly highlighting the cyber risks and data-security dangers of feeding sensitive information into public AI tools.

Beyond training, Lloyds addresses financial and structural barriers through its broader industry ecosystem. Alongside its ongoing work with the SME Digital Adoption Taskforce, the bank provides targeted backing to advanced tech hubs like the MTC. This ensures SMEs have a direct pipeline to the expert support, technical advice, and foundational funding pathways needed to adopt AI-powered digital tools confidently.

Further ahead, Rodrigues sees AI combining with other emerging technologies, such as tokenisation – the process of representing real-world assets digitally on a blockchain – to help address long-standing financial challenges.

“Combining AI with tokenisation unlocks huge knowledge about the patterns of how money moves,” he says. While it has not fully reached SMEs yet, the implications for speeding up trade finance and cross-border payments are significant.

The competitive advantage of acting now

For now, Rodrigues’ message to SMEs is a simple one: start. Those who embrace the technology, however modestly, will be far better placed than those that do not.

“It’s transforming everything: every industry, every business’ way to both drive revenue and remove costs. It affects both sides of the balance sheet.”

Ultimately, the true return on investment for Britain’s SMEs may not just be measured in pounds and pence, but in time. By taking care of the repetitive, administrative heavy lifting, AI is giving business owners and their teams the freedom to return to the things that matter most: their core expertise, human creativity and building lasting relationships with their customers.

*Lloyds Business and Commercial Banking Customer 360 Survey

All lending is subject to status. Eligibility criteria apply. Lloyds Bank data correct as of April 2026. Figure includes clubs, charities and societies.

*Lloyds and Lloyds Bank are trading names of Lloyds Bank plc. Registered Office: 25 Gresham Street, London EC2V 7HN. Registered in England and Wales no. 2065. Telephone: 0207 626 1500. *

Authorised by the Prudential Regulation Authority and regulated by the Financial Conduct Authority and the Prudential Regulation Authority under Registration Number 119278.