The Forgotten Blueprint for Modern AI

A few weeks ago, I spent a day at Bletchley Park. Like many people, I arrived expecting to learn about machines. I expected to see the history of codebreaking, early computing and the technology that helped shape the modern world. What I did not expect was to leave thinking about Salesforce, artificial intelligence and business enablement. Yet that is exactly what happened. As I walked through the exhibits, listened to the stories and learned more about the people who worked there, I found myself reflecting on a question that organisations everywhere are currently wrestling with. What actually makes AI work? At first glance, the answer seems obvious. Powerful algorithms. Advanced computing. Sophisticated platforms. Vast quantities of data. These are certainly part of the story. But they are not the whole story. Because the more I learned about Bletchley Park, the more I realised that its success was not primarily a story about machines. It was a story about people. Thousands of people. Mathematicians, linguists, analysts, clerks and specialists working together to solve one of the most complex information challenges of their time. Their job was not simply to process data. Their job was to take huge volumes of chaotic, incomplete and often unreliable information and turn it into something meaningful. In many ways, that challenge feels surprisingly familiar. Today, organisations face their own versions of information overload. Data flows through CRM systems, service platforms, reporting tools and analytics environments at an extraordinary rate. Leaders are expected to make decisions faster than ever before. Artificial intelligence promises to help by interpreting information, identifying patterns and generating recommendations. The technology has changed dramatically. The underlying challenge has not. We are still trying to transform information into understanding. And that is why the lessons of Bletchley Park remain remarkably relevant. One of the most interesting aspects of the Bletchley story is that success depended on far more than intelligence. It depended on discipline. The people working there did not simply sit down and start guessing. They followed structured processes. They worked within clearly defined frameworks. They applied consistent methods. They validated information rigorously. There was a shared understanding of how work should be performed. There had to be. When the consequences of getting something wrong are significant, consistency becomes essential. A different interpretation by one individual could create confusion for everyone else. Inconsistency was not merely inefficient. It introduced risk. This is the point that stayed with me long after my visit. Because when I look at many organisations today, particularly those preparing for AI adoption, I often see the opposite. Different teams use data differently. Definitions vary. Processes are interpreted in multiple ways. Information is entered inconsistently. Standards exist on paper but not always in practice. The environment lacks the discipline that meaningful analysis depends upon. Yet many organisations still expect AI to generate clarity from that complexity. That expectation deserves closer examination. The current conversation around artificial intelligence often focuses on what the technology can do. We hear about automation, productivity gains, predictive insights and intelligent recommendations. In the Salesforce ecosystem, conversations increasingly revolve around platforms such as Agentforce and the opportunities they create. These capabilities are exciting. However, there is a tendency to skip an important question. Is the organisation actually ready for them? Readiness is not primarily about technology. Most organisations already possess more technology than they fully utilise. The real question is whether the underlying foundations exist. Is the data reliable? Are processes followed consistently? Do users share a common understanding of what good looks like? Do people understand the information they are creating? Without these foundations, AI has very little to work with. This is where many assumptions begin to unravel. One of the most persistent myths surrounding artificial intelligence is the belief that it will somehow fix poor-quality information. Organisations struggling with inconsistent data often see AI as a future solution rather than recognising it as a force multiplier. The reality is far less forgiving. AI does not remove inconsistency. It learns from it. If users interpret opportunity stages differently, AI learns those differences. If records are incomplete, AI works with incomplete information. If processes are loosely followed, AI identifies patterns within that inconsistency and treats them as meaningful. In other words, artificial intelligence amplifies what already exists. Good foundations become more valuable. Poor foundations become more visible. This became particularly clear to me recently when I noticed a growing number of job advertisements for AI training specialists. At first glance, the role sounded highly technical and futuristic. Yet when I looked more closely at the responsibilities, the work appeared surprisingly familiar. The role involved classifying information. Labelling data. Following strict guidelines. Applying consistent standards. Evaluating outputs against defined criteria. In short, it involved creating order. The work was structured, disciplined and methodical. It was fundamentally human. And the more I thought about it, the more it reminded me of what Salesforce users should already be doing every day. Every time a user updates a record, selects a field value, logs an activity or records information about a customer, they are helping define what good data looks like. They are creating patterns. They are establishing consistency. They are training the system. The difference is that organisations rarely describe it that way. Instead, data entry is often treated as administration. Something users must do because the system requires it. Yet in an AI-enabled environment, those actions take on a much greater significance. They become the foundation upon which future recommendations, insights and automations are built. This creates a challenge that many organisations have not fully recognised. Historically, Salesforce training focused on navigation. Users learned where to click, how to update records and how to complete processes. Success was measured by whether someone could use the platform. In an AI-driven environment, that is no longer enough. Users increasingly need judgement. They need to understand why recommendations are being made. They need to evaluate whether suggestions make sense. They need to recognise when information appears inaccurate. They need the confidence to challenge outputs when necessary. This represents a significant shift. The conversation is no longer simply about system usage. It is about decision-making. And that is why enablement is becoming more important than ever. Not because organisations need more training for the sake of training. But because users need a deeper understanding of how information flows through the organisation, how data influences outcomes and how artificial intelligence depends on the quality of what people contribute. When users understand this connection, their relationship with Salesforce changes. They stop seeing it as a system they are required to use. They start seeing it as a system they actively contribute to. That distinction matters. Because the future of AI is not determined solely by the quality of the technology. It is determined by the quality of the environment surrounding it. This brings us back to Bletchley Park. The people who worked there did not rely on machines to create order from chaos. They built the discipline, consistency and understanding that allowed machines to become useful. The technology was powerful because the foundations were strong. Today, many organisations are attempting to reverse that sequence. They are introducing increasingly powerful tools into environments that still struggle with consistency, clarity and capability. They hope the technology will create the order that does not yet exist. History suggests otherwise. The lesson from Bletchley Park is not that technology is unimportant. It is that technology works best when it sits on top of strong human foundations. Before intelligence comes discipline. Before automation comes consistency. Before artificial intelligence comes enablement. The organisations that understand this will be the ones that gain the greatest value from the next generation of technology. Not because they adopted AI first. But because they prepared their people first. And in the end, that may be the most important lesson Bletchley Park still has to teach us.