Most organisations are already using AI. Very few are ready for it.

Any number of surveys find that firms are already using AI in some form. But most of that usage is concentrated in low-value productivity tasks: drafting emails, summarising documents, generating first passes at presentations. The tools are being used, but they are being used without a strategy, without guardrails, and often without the organisation even knowing it is happening.

This is not an adoption problem. It is a readiness problem. The technology arrived faster than most organisations could prepare for it, and the result is a widening gap between what people are doing with AI and what the organisation has in place to support, govern and direct that activity.

The major AI providers are in an arms race. New models, new capabilities and new vendor offerings are arriving at a pace that makes it a full-time job just to keep track. The temptation for organisations is to try to keep up: to chase each new release, pilot each new tool, and respond to every headline.

But chasing the technology is not the same as making progress. Global investment in AI is estimated at around 500 billion dollars, yet only around 5% of that has translated into realised business benefits. The problem is not a lack of spending or ambition. It is that organisations are investing in tools and pilots without first being clear about what they are trying to achieve, what they will and will not permit, and how they will know whether any of it is working. And we should be cautious of research showing low levels of benefit – it it’s not a reason to cool on AI, but more a reason to focus on the outcomes you want.

There is a useful way to think about the risks that organisations are running. The first is the verification gap: the human capacity to check AI outputs for accuracy. AI is confident, fluent and frequently wrong. If the people using it do not have the skills or the time to verify what it produces, errors will reach customers, regulators and the market.

The second is the thinking gap: the human capacity to apply critical judgement to what AI generates. When people start accepting AI outputs without interrogating them, the quality of decision-making degrades, even if the surface quality of the work improves.

The third is the confidence gap: the human capacity to understand and stand behind material that AI has helped produce. If someone cannot explain or defend what is in a document they have submitted, the fact that an AI wrote most of it is not a defence. There is no liability for the AI provider. The accountability sits entirely with the person and the organisation.

Readiness is not about having a perfect AI strategy before anyone touches a chatbot. It is about putting three practical foundations in place. The first is strategic focus: a clear view of which problems AI should help you solve, so that effort is directed rather than scattered. The second is a set of guardrails: a short, practical AI policy that gives people permission to use AI while setting boundaries on what is and is not acceptable. The third is governance: a cross-functional group, whether you call it an AI council or something less formal, that takes responsibility for monitoring, interpreting and acting on the changes happening around you.

None of these require large budgets or long timescales. They require attention and leadership. The organisations making real progress are not the ones with the most advanced technology. They are the ones that have decided what they are trying to do, set boundaries for doing it safely, and created a forum for staying on top of a landscape that will not slow down for anyone.

If your organisation is already using AI but has not yet put these foundations in place, the gap is growing every day. The practical starting point is a readiness assessment: an honest look at where AI is being used, what governance exists, how well-equipped your people are to use it responsibly, and where the risks sit.

Red Olive helps organisations build the data foundations, governance and analytical capability needed to adopt AI in a way that is purposeful, safe and measurable. If you want to close the gap between what your people are already doing and what your organisation is ready for, we can help.