AI is changing how we work. The smartphone era offers a warning: technology can become embedded in working life before organisations have decided how it should be used. Smartphones changed not only how people worked, but what work came to demand of them. Convenience gradually became an expectation, and constant availability became the norm before its costs were fully understood. AI will herald a more profound transformation: not simply making existing tasks faster, but changing how work is organised, valued and understood. Whether that transformation proves ultimately beneficial will depend not only on what the technology can do, but on how we choose to use it.
The smartphone precedent
Smartphones fundamentally altered how people communicate, work, socialise (or not), shop, learn and even date. They enabled faster communication and more flexible ways of working, but the resulting “always-on” culture blurred the boundaries between professional and personal life. Constant connectivity was also associated with burnout, digital fatigue, distraction and work-related stress.
The precedent is not that smartphones and AI are the same. It is that smartphones became embedded in working life before most organisations had decided what good use looked like; by the time they did, the habits were already established.
AI is different, however. For many people, smartphones changed where and when work happened more than they changed the work itself. They made existing tasks portable and accelerated communication, but the underlying job largely remained the same. AI can perform parts of the work itself, potentially changing not only working patterns but the division of labour between people and machines.
How AI changes the finance function
At VantagePoint, we build lean, automated systems and processes for the finance function. The impact of AI on finance is therefore practical for us, not theoretical.
AI tools can now draft reports, summarise meetings, analyse data, generate code and support decision-making. Combined with established workflow and rules-based automation, they can extend routine finance processes such as invoice processing, reconciliations, month-end close and reporting, for example by classifying exceptions, drafting commentary and identifying anomalies.
AI can also support forecasting, budgeting and scenario planning by processing large volumes of internal data faster than manual workflows. Combining appropriately governed internal and external data can enable more responsive financial planning and analysis. In risk management, fraud detection, compliance and cash flow monitoring, AI can help surface anomalies, trends and emerging issues for human review.
As these tools mature, finance teams will spend less time producing information and more time interpreting it, translating analysis into business action, and working with technology, operations, and leadership teams. AI is likely to reshape many finance roles more than simply replace them, making the function more efficient, predictive and strategic; in doing so, it will demand more of the people tasked with delivering smarter decisions and better insight.
The human risks of automation
When AI can do much of the manual, tedious work, it changes the texture of the working day. Routine tasks can provide periods of respite; automate them all away and the day becomes more intellectually satisfying but also more cognitively demanding and exhausting.
As more routine tasks are automated, AI fluency will become increasingly important. Finance professionals will need stronger data-interpretation skills and the judgement to know when to trust an output, challenge it or refine the prompt, and how to spot errors and bias. Some roles will also require expertise in automation, model validation and governance. Organisations will need to invest accordingly in reskilling and workforce development.
Accounting fundamentals, professional scepticism and a strong grasp of internal controls will still matter. They will increasingly sit alongside commercial storytelling, scenario thinking and risk judgement. Human skills such as emotional intelligence, leadership, ethical judgement, creativity and relationship management may become more valuable as routine cognitive work is automated. These are not new requirements for finance; AI raises their importance because producing an answer will become easier than judging it or deciding what to do with it.
AI also gives employees a new benchmark against which to judge themselves: colleagues who adapt faster, and the machines themselves. If employers then use the same technology to monitor performance, a tool intended to help can start to feel like surveillance. That pressure will be difficult to contain as long as employees remain unsure what the technology means for their own jobs.
Cognitive dependence is another risk. As people delegate more analysis, writing, planning and problem-solving to AI, we may outsource more of our mental effort to it. Most of us no longer read maps; we rely on sat-nav. How many of us could navigate with a paper map today? The same risk applies in finance: if people stop practising the analysis and controls that underpin an output, fewer will be able to recognise when a polished answer is wrong.
AI-generated communication also threatens to reduce authentic human interaction. AI-generated reports may be faster to produce but reduce the need for collaboration and, with it, the personal connection between colleagues.
The choices finance leaders must make
If employers wait for AI's effects to become obvious before setting boundaries, damaging habits and expectations may already be part of the culture.
The long-term impact on workplace culture will depend largely on whether organisations use AI to empower employees or simply extract more productivity from them. Used purely to maximise output, AI may create workplaces that are technically productive but psychologically unsustainable. Finance leaders will be challenged to deliver more value, likely with fewer people, while improving their teams' lives.
How organisations use AI's productivity gains will be central. They can return them to employees, invest them in deeper analysis, learning and business partnering, convert them into higher output, or use them to support a smaller team. The technology does not make that decision; management does.
For each use case, finance leaders should define what successful adoption looks like before deployment. They need to decide which outputs require human review, which skills the team must retain, what monitoring is proportionate and when employees should be able to switch off.
The smartphone era should have taught us that technical progress and a better quality of life are not the same thing. Productivity can rise even as wellbeing deteriorates. That is the danger employers should bear in mind as they decide how to deploy AI.
Turning choices into a plan
At VantagePoint, we help finance teams identify where AI can create value and introduce it with appropriate governance. The goal is not automation for its own sake, but a stronger finance function in which better technology supports human judgement rather than substituting for it.
That is the thinking behind our CFO Labs offering. Before each one-day workshop, we use surveys and interviews to establish the finance function's current state and readiness for change. The session explores how AI may reshape finance, with collaborative exercises that help leaders decide what they want AI to do for their organisation.
From there, we identify where AI could reduce manual work, improve workflows or make information easier to find and use. The result is a practical roadmap covering priority initiatives, integration steps, enablers and risks, which we can help finance teams implement across the function. It identifies what to automate, where human review still matters and which skills the team needs to develop.
These decisions are worth making now. Finance leaders have the chance to approach AI differently: deciding what good use looks like before the technology and the habits around it become embedded in the working day.
.avif)
.avif)
