Authentic Intelligence: rooting AI in daily lived experiences

This is a guest blog post by Aaron Harris, CTO, Sage.

Every week brings another AI announcement promising significant improvements to business operations. Yet for many IT leaders, the reality feels more like déjà vu than breakthrough. Bold headlines claim that new AI solutions will bring dramatic enhancements, but when they arrive, too often are they met with a collective “so what” at best and frustrated annoyance at worst.

Businesses and accountants are stuck in a cycle of massive promise followed by underwhelming delivery, because too often, AI is inserted into applications and products without real consideration about how helpful it actually is.

It seems that a lot of developers have forgotten the lessons from other infamous stumbles. The tragic life of Clippy, Microsoft Word’s reviled virtual assistant, is a perfect case study. If you’ve ever wrestled with Microsoft Word in the early 2000s, you probably remember that overly enthusiastic googly-eyed paperclip. The original intelligent virtual assistant, its trademark, “It looks like you’re writing a letter. Would you like some help?” was meant to be helpful, but as users became more proficient, it became irritatingly intrusive. The fundamental problem was that Clippy didn’t understand user intent well enough. In Clippy’s universe, everything was a letter to a loved one.

New solutions for real problems

But Clippy deserves some credit. It walked so that modern AI assistants could run. While Clippy could only suggest, today’s AI assistants can and should be so much more authentic, trained on the real experiences of its users and insights of their professions, to anticipate and support as well as advise.

That’s the promise of agentic AI: intelligent systems that don’t just respond to prompts, but monitor, interpret, and act independently across many tasks and workflows – a genuine leap forward for industries like finance.

It’s transforming entire professions, like accounting, in the process. When trained on decades of industry-specific data and connected with agents across all areas of accountancy – from compliance, payroll and cashflow – AI starts to truly understand the needs of accountants and the businesses they serve. It frees them up from tasks that drain productivity, so they can focus on growth.

Consider the UK’s Making Tax Digital (MTD) mandates, which are making it compulsory for businesses to use a digital system to keep tax records, submit tax data and make payments. This marks a significant shift in how accountants and their clients manage tax obligations, adding pressure on practices already stretched for capacity. AI agents can relieve that burden by automatically sorting clients based on complexity, setting up tasks and reminders, chasing documents, and flagging potential issues early. It helps accountancy practices stay compliant with MTD, cut down on manual work, and free up more time to support their clients.

Building trust through transparency

Of course, the benefits only matter if the technology is trusted. Trust remains a huge barrier, with our own research showing that there is a direct correlation between trust and adoption. While (94%) of SMBs already using AI report seeing benefits, the majority

(70%) have yet to fully adopt the technology, with this gap caused by confidence rather than capability.

Especially in finance, trust isn’t optional, it is everything.

Finance teams work under strict audit requirements. Every number must be verifiable, which means a rogue AI agent could cause serious damage. A rogue agent that hallucinates and fabricates data or misclassifies invoices could cause chaos in company accounts. Without strong safeguards or a clear “stop” function, the fallout could be detrimental.

Businesses deserve to know how their AI technology works, how their data is used and what safeguards are in place. That’s why it’s so important to make sure AI development is transparent and responsible by design, like we’ve done with our AI Trust Label. More companies should be making visible pledges to support customers in understanding how AI impacts them without needing technical expertise.

From hype to authenticity

The businesses seeing genuine AI value today aren’t chasing the latest features. They’re implementing solutions that understand their industry deeply, operate transparently, and earn trust through consistent, reliable performance.

For IT leaders evaluating AI investments, the questions should shift from “What can this AI do?” to “How does this AI understand our specific business context?” and “Can our teams trust it to make decisions autonomously?”

That’s how they’ll move beyond the hype toward AI that’s built with them, for them, so they can focus on what really matters.