Crowd Minds

To summarise the core imperative of this article in one sentence: “Let the data flow!” AI tools drive productivity, AI agents automate processes. Put the discussion about the AI bubble aside — we all agree that AI has significantly changed the way we work, and it continues to find applications in our daily lives. Organisations are searching for the best strategy to adopt AI in order to enhance competitiveness, productivity, and creativity. I have been fortunate to experience different organisational structures, rules, and mindsets throughout my career. These experiences gradually led me to a personal conclusion about how organisations must operate in the age of AI.

• Data should be free to use. Protect only those which must be restricted.
• Ensure availability of tools — don’t select the best one for your people, let the people decide what they need.
• Automate processes and enforce policies so that people can work independently within the realm of what is possible.

AI readiness of a company can be measured by the speed at which data flows from source to destination. In this context, we are not referring to traditional technical pipelines such as operational databases feeding data lakes. Instead, AI readiness reflects how quickly individuals inside the organisation can retrieve the information they need and act on it. Consider a few simple questions: how long does it take for an employee to gain access to a dataset that could help them solve a problem? How long does it take to install or experiment with a tool they need for their work? How often do employees depend on other people simply to access information or complete a step in a process?

In many organisations, data exists in what could be described as sealed boxes. The data technically exists somewhere in the infrastructure, yet in practice it is difficult to access. Organisational silos, unclear ownership, and long approval chains create barriers that slow down the movement of information. In traditional environments this friction reduces efficiency; in an AI-enabled environment it becomes a structural limitation. AI systems amplify the value of accessible data, but they cannot create value from data that remains locked behind organisational barriers.

Opening these boxes does not mean abandoning governance or security. It means reversing the default assumption about access. Instead of assuming that data should be restricted unless explicitly permitted, organisations should assume that data can be used unless there is a clear reason to restrict it. When access becomes the norm rather than the exception, the analytical capacity of the organisation increases dramatically. Employees begin to explore information across domains, experiment with new ideas, and combine datasets in ways that central planning could never fully anticipate.

The same philosophy applies to analytical tools. Knowledge workers today need the ability to retrieve data, analyse it, model it, and automate workflows. If every new tool requires a lengthy approval process, experimentation slows down and opportunities are lost. Centralised tool selection often aims to optimise standardisation, but it can unintentionally limit innovation. In reality, the best tool often depends on the specific problem being solved. Allowing individuals and teams to choose the tools that fit their needs encourages experimentation and accelerates learning.

Once data becomes accessible and analytical tools are available, the next step is automation. Many organisational processes today still rely on manual coordination between people: forwarding information, requesting confirmations, or manually triggering the next step in a workflow. Each handover introduces delays and increases dependency between individuals. Automation — supported by AI agents and workflow systems — can remove many of these intermediate steps and allow processes to progress automatically.

The objective is not to remove people from processes, but to minimise unnecessary interdependence. When individuals can independently access information, analyse it, and trigger the next step in a process, the organisation becomes significantly more responsive. AI then acts as a multiplier rather than a bottleneck. Agents assist with analysis, automation handles repetitive operations, and people can focus on decisions that require judgement, creativity, and strategic thinking.

From this perspective, AI readiness is less about deploying the most advanced models and more about removing friction from the flow of information and action. The organisations that will succeed in the long run are not necessarily those with the most sophisticated algorithms, but those that allow data, tools, and automated processes to move freely across internal boundaries. When information flows, experimentation becomes natural, decisions accelerate, and the organisation becomes capable of adapting continuously in an AI-driven world.