All our tomorrows: the path and impact of AI

I come at AI from quite a blended background. I was  a database administrator (DBA), and transitioned into software development. I saw the potential of AI and now focus on delivering intelligent AI-powered solutions to drive business transformation for both small and large organisations.

The original challenge for that transformation was around scale — how to support growing demand for complex services (particularly in care and compliance-heavy sectors) without proportionally increasing staff or costs.

Typically, businesses have had to address these situations with manpower.  However, this quickly becomes less than idea as a skilled, even professional workforce gets bogged down with manual, repetitive, low-value tasks. AI offers a way to unlock latent productivity and allow skilled professionals to focus on higher-impact work.

There have already been tangible benefits. Cap Gemini figures have shown that early adopters of Generative AI have seen improvements in operational efficiency and customer experience. On average, organizations realized a 7.8% improvement in productivity and a 6.7% improvement in customer engagement and satisfaction, between 2023 and 2024.

It is therefore unsurprising that 80% of organizations have increased their investment in Generative AI from 2023 to 2024. Larger businesses lead the charge, but overall, nearly a quarter (24%) of organisations have integrated AI into their operations.  For context, that compares to just 6% reported the year prior[1].

 

The issue of today

But despite this huge promise, there are still significant limitations. Generative AI, in our experience, is currently around 70% accurate and that is an issue that needs addressing.

Gen AI often requires deep prompt engineering expertise to generate reliable outputs, and even then, human oversight remains essential.

This is particularly evident in development environments, where business can use tools like VIBE coding for speed, but still encounter linting issues and errors that demand either re-processing or manual correction.

In essence, Gen AI is still not plug-and-play. Businesses must understand the importance of human-in-the-loop processes for quality assurance – and that means there is a limit to efficiency improvements, unless the AI can be put to work elsewhere.

 

The AI of tomorrow

In this case, ‘elsewhere’ means intelligent systems that can operate autonomously, interact with tools, and learn through reinforcement and machine learning.  This is the emerging shift from Generative towards Agentic AI.

This evolution is easily underestimated. Widely adopted Generative AI tools offer quick wins but limited depth. Agentic AI opens the door to long-term business transformation by acting like virtual employees capable of executing entire workflows.

This will be the next technological payload that forward-thinking companies will adopt.

Just as ChatGPT revolutionised access to intelligent assistance, Agentic AI is now poised to reshape operational models. AI will not just generate content, but take action, learn, and improve outcomes over time.

The promise of Agentic AI has led to organisations eager to adopt the technology. Cap Gemini research has shown that 82% intend to integrate agents within three years. The report found an established level of trust in AI agents for specific tasks, such as generating work emails, coding, and data analysis.[2]

However, a healthy scepticism remains: the same research found organisations recognise the need to establish guardrails to validate AI-made decisions.  It is clear that transparency and accountability remain key concerns, despite Agentic AI already being worth more than £156m in the UK alone[3].

Across both Generative – and Agentic AI (as well as the developing Inference models) it is undeniable that AI will  shape the workplace of the future.

 

The office of tomorrow

There has been a remarkably varied set of responses to AI. At one end, recent research 2024 found almost a third of UK businesses were scared to implement the tech[4]. Elsewhere, some futurists have warned it will cause to re-architect entire services, building them around AI.

However, it is likely that the middle ground will be the majority, where will change jobs rather than replace them outright. While investors may be enticed by cost-saving implications, most businesses advocate for AI as a productivity and profit enabler, not a job killer. They draw parallels with tools like Zapier, which long before the AI boom had already automated many HR and admin tasks.

The good news is that while the move toward automation and augmentation is inevitable, but this should empower employees rather than replace them.

That inevitability, and the pervasiveness of AI means that the technology industry – and wider business – needs to address the ethical implications of AI, particularly regarding its influence on younger generations.

There’s a growing need for dedicated AI education in schools — not just in how to use these tools, but in understanding their implications. Firstly, it is clear that developing prompt engineering skills is a clear advantage.

But, at a deeper level, for young people with learning disabilities, AI tools — particularly in the edtech space — can offer transformative support that improves access to education and tailored learning paths.

At its most sophisticated, there is then the use of AI-generated avatars.  But educators and businesses alike must be careful and deliberate in their deployment, ensuring they augment, not impersonate, human interaction.

 

Conclusion: the path to tomorrow

It is a cliché, but no less true for it, that AI is the shape of the future. To know – or indeed influence – the path of AI evolution is the closest thing to a crystal ball that businesses have ever had.

That path will see AI go beyond being just as a tool for automation, and into becoming a catalyst for reshaping productivity, creativity, and collaboration. The key is to apply it responsibly, with clear intent and human oversight, and always with a view to empowering people — not sidelining them.

 

[1] See Generative-AI-in-Organizations-Refresh_25112024.pdf

[2] See Generative-AI-in-Organizations-Refresh_25112024.pdf

[3] See UK Enterprise Agentic AI Market Size & Outlook, 2030

[4] See 1 in 3 UK Small Businesses Held Back by AI Fears | Onrec