UK businesses risk losing billions of pounds annually through poorly implemented artificial intelligence strategies, according to new analysis from training provider The Coders Guild, which warns that weak governance, inadequate staff training and growing “shadow AI” use are undermining productivity gains.
The organisation estimates that if just a quarter of the UK’s 5.5 million businesses are using AI inefficiently — and each loses between £5,000 and £10,000 a year through wasted time, duplication and missed automation opportunities — the cost to the wider economy could range from £6.8bn to £13.7bn annually.
The warning comes amid surging business interest in AI tools, but limited evidence that firms are putting the right structures in place to deploy them effectively.
Training Gap Leaves Businesses Exposed
According to data cited by The Coders Guild, 33% of UK employers have expressed interest in AI training, yet only 11% have delivered any form of AI training over the past year. Among small and medium-sized enterprises, fewer than 12% have invested in structured AI training programmes.
The disparity has created what experts describe as a “curiosity-to-commitment gap”, with firms exploring AI’s potential but failing to invest in the capability needed to use it productively.
Crispin Read, founder of The Coders Guild, said many businesses are adopting AI without the governance or operational framework needed to support it.
“Without clear strategy or training, many organisations are relying on free, off-the-shelf AI tools with little oversight or integration into business processes. As a result, teams are spending increasing amounts of time validating unreliable outputs, often duplicating work across departments and making decisions based on AI-generated responses that haven’t been properly scrutinised.”
He added that this is creating hidden inefficiencies at the same time as businesses expose themselves to data security and compliance risks.
Leadership Confidence Emerging as Key Barrier
The report suggests the main barrier to effective AI adoption is not employee resistance, but lack of confidence among senior leadership.
Only 34% of leadership teams say they feel confident identifying AI opportunities within their business, while fewer than 30% of small businesses report confidence in implementing AI effectively. Around 60% of firms say their employees lack adequate AI training.
Read said leadership capability is central to unlocking wider adoption.
“Only a third of senior leadership teams feel confident identifying AI opportunities – that’s the number that explains all the others. You can’t commission the right training if you don’t understand the problem. You can’t ask the right questions of your suppliers, or your tech team, or the vendors trying to sell you tools. AI literacy at leadership level isn’t a nice-to-have – it’s what determines whether anything else gets done.”
Shadow AI Use Raises Governance Concerns
The Coders Guild also flagged the rise of “shadow AI” — where employees use AI tools informally without company oversight — as an increasingly urgent concern for employers.
Read said many firms underestimate the extent to which staff are already using generative AI in day-to-day operations.
“The shadow AI finding is the one I’d pay most attention to. Employees already use AI tools informally, without any training framework or policy behind it. That’s not a future risk – it’s already happening in most organisations. And businesses often don’t realise until something goes wrong with a client’s data, or a decision gets made on the back of an output nobody’s thought to question. That’s a governance gap more than a technology gap, and it’s one structured training can actually close.”
Businesses Urged to Measure Hidden Costs
Read said many businesses remain hesitant to invest in AI because they struggle to quantify the return on investment, despite potentially incurring significant unseen losses from outdated processes.
“Most businesses that tell me they can’t justify the investment haven’t tried to measure what they’re currently losing. The waste is usually visible pretty quickly – manual processes that exist out of habit, hours spent on tasks that a well-configured tool could handle in minutes, decisions made without data that’s sitting right there. You can’t see the bottom line benefit until you start looking for it, and that’s what good AI training actually teaches you to do.”
Research cited by the organisation suggests businesses that implement AI effectively could achieve a 30% to 40% productivity advantage by 2028.
New Apprenticeship Aims to Address Skills Shortage
In response to the widening digital skills gap, The Coders Guild has partnered with Skills England to launch a new Level 4 AI and Automation Practitioner apprenticeship.
The 18-month programme is designed to help businesses either recruit new talent or upskill existing employees in AI and automation, with practical training focused on integrating AI into operational workflows.
The apprenticeship also covers ethical AI use, data protection and compliance to support more responsible deployment of the technology.
Risk of a Two-Speed Economy
The Coders Guild warned the UK faces the prospect of a two-speed economy, with AI-enabled firms significantly outperforming businesses that fail to build internal capability.
Read said many business leaders recognise the urgency but remain uncertain about where to begin.
“The numbers here don’t surprise me – they reflect almost every conversation I have with business leaders right now. People know they need to do something about AI. They’re searching for it, talking about it, worrying about it. The bit that’s missing is a practical first step that fits how their business actually works. For a lot of companies, that isn’t just hiring someone new – it’s giving the people who already understand the business the skills to improve it using AI.”
“Friendly zombie fanatic. Analyst. Coffee buff. Professional music specialist. Communicator.”
