Whether you love it or hate it, artificial intelligence (AI) is everywhere, and it’s here to stay.

While it isn’t a new concept, the hype around AI truly began to soar in 2023, when advancements in generative AI, particularly models like OpenAI’s ChatGPT-4, started making waves in content creation, answering questions and automating complex tasks.

Businesses quickly caught on to its popularity, with 39% of UK organisations adopting AI technologies in 2023 – a breakout year that saw it move from a niche tool to a mainstream business solution.

But despite the buzz around AI, it hasn’t worked out for everyone.

Research has revealed that over a third of UK businesses that have tried implementing AI solutions in the last 12 months have seen these projects fail. But, why has that been the case?

AI for business – expectations vs reality

AI became democratised in 2023. It wasn’t just for tech experts anymore – it was something anyone could pick up and use.

But it wasn’t all about answering quick questions, creating funky filters, generating wacky images or writing on-the-spot poems. Businesses were also able to make use of its capabilities, particularly to automate repetitive tasks, improve customer service and streamline operations.

Many organisations were optimistic about AI’s impressive functionality, particularly as 2024 began. A survey conducted by Startups revealed that businesses that were preparing themselves for AI disruption felt better positioned for the new year, with 14% of respondents believing that investing in AI should be their priority in 2024. Companies within the Technology, Fintech and Healthcare sectors were also reported as the top industries for AI adoption. 

Yet despite the optimism, implementing AI has proven to be harder than expected. According to research by OneAdvanced, 52% of UK businesses have attempted to integrate AI into their everyday operations, but 36% of leaders reported that those efforts have failed.

While the IT industry saw the best success, with just a 16% failure rate, other sectors have struggled more. This includes 30% of GP practices and 29% of finance businesses that reported problems in the survey.

AI-specific skills are lacking

So, why have so many businesses struggled to implement AI successfully?

OneAdvanced’s research suggests that a lack of AI-specific skills and experience is to blame. It found that while 85% of business leaders acknowledged the benefits of AI solutions, just 28% believed that their workforce had sufficient skills to enable the use of AI effectively.

Simon Walsh, CEO of OneAdvanced, cited the UK’s employee engagement crisis and leaders failing to skill up their workforce as the primary reason for this issue.

“Productivity in the UK is at an all-time low, with leaders’ inability to skill up and embrace AI, risking them lagging behind their global counterparts,” he said. “Businesses are being held back by a lack of expertise, which often leads to a gold rush for certain skills to be hired or third parties to supplement the gaps.

“Collectively, we simply cannot afford to wait for the right people to become available, or the country risks losing significant ground against the global competition.”

Moreover, while AI roles were reported to be the highest-paid in the tech industry, more than two-thirds (68%) of business leaders said that they’ve struggled to attract the right talent to manage their AI tools effectively.

Businesses were quick to jump on the AI “trend”

Given AI’s surge in popularity over the past year, it’s no surprise that a lot of businesses have jumped on the bandwagon.

The problem is that AI projects can fail because of a lack of clear business objectives, and as a result, may fail to deliver any measurable value or align with the company’s strategic goals. This has been the case for up to 80% of AI projects in 2023, as they weren’t able to achieve their full potential or meet business goals.

However, this isn’t always down to the decisions of founders or business owners. Rather, it can come from pressure to adopt emerging technologies to stay competitive, meet customer expectations or keep up with industry trends – even when the business might not be fully ready for it.

Research conducted this month from Startups found that 82% of UK small businesses are experiencing pressure to adopt technologies like AI. But despite this push, rushing into AI adoption without proper preparation can lead to wasted resources, unmet expectations and even financial losses.

Poor data quality is a major obstacle

AI systems rely heavily on data. But, poor quality, unstructured or inaccessible data can lead to substandard models. As a result, this can lead to unreliable predictions or insights, making the system untrustworthy.

According to a study reported by Digitalisation World, 60% of organisations say that AI is a key influence on data programmes. However, just 12% reported that their data is sufficient and accessible enough for effective implementation. The lack of good-quality data itself was also reported to be a challenge for 64% of organisations, in turn holding them back from being able to fully utilise AI’s potential for decision-making, automation and strategic insights.

Josh Rogers, CEO at Precisely, commented: “To fully capitalise on the business benefits of analytics and AI, organisations need to invest in data integrity. Establishing a foundation of accurate, consistent, and contextual data can serve to help them make informed decisions with confidence and truly realise the value of their AI initiatives.”

Can AI work for your business?

Despite the large number of failed AI projects, 88% of UK businesses are prioritising preparing for AI adoption for 2025. Our own research has also found that 50% of businesses cited improving efficiency and productivity as the most influential factor in their technology adoption decisions.

If you’re looking to implement AI technology into your own business, here are a few strategies that can help: