AI has forked in two clear benefit streams.

• Personal AI – the tools that we use as people – Co-Pilots, Coda, ChatGPT, NotebookLM
• Business AI – the tools that are introduced by a business to boost productivity of staff and processes and systems, and capture quality data.

95% of AI Pilots fail

New research from MIT called ‘The Gen AI Divide State of AI in Business 2025’, examining 300 public implementations, reveals what researchers call the “Gen AI Divide”: a split between the small minority achieving substantial value and the majority trapped in failed pilot programmes.

The primary root cause identified by the MIT “GenAI Divide” report for the 95% failure rate of AI pilots is that organisations treat AI as a standalone tool or isolated experiment, rather than deeply integrating it into business processes, data infrastructure, and workflows. This disconnect means pilots often fail to scale, remain detached from real operational needs, and deliver little measurable business value.​

Specifically, the report and supporting analysis highlight these compounding factors:

• Lack of clear business objectives and integration with existing systems and processes.​
• Weak data foundations and fragmented or poor-quality organizational data.​
• Insufficient investment in change management and organisational readiness, including lack of leadership sponsorship and user buy-in.​
• AI initiatives that are hype-driven, poorly coordinated, and not aligned with tangible ROI goals.​

In summary, AI pilots predominantly fail not because the technology is flawed, but because enterprises are unprepared to embed and scale AI throughout the organization, both technically and culturally.

Another cause for concern is many businesses are being seduced by technologies and terms with a desire to be seen to be introducing AI and therefore bolster their digital credentials.

The reality is that AI will best serve a business when it is being implemented as part of the strategic objectives and in line with clearly identified business needs.

The most simple test is to look at your Business Strategy.

Are all your KPIs measured and reported using metrics derived from your data-sets without any human curation?

• If the answer is “Yes, all our KPIs are measured and reported on using our own data sets without any human curation“. then the chances are, your data is AI ready (or at the very least, well on the way).
• If the answer is “No, not all our KPIs are measured and reported using our own data sets without any human curation“, then you are unlikely to be AI ready.

All your standard reports, both financial and others, need to be driven from your data without any human curation.

If you want a more comprehensive AI Readiness Assessment, please use this free tool.

Underpinning your AI readiness will be your Information Framework.

Do you have an Information Framework driven by and endorsed by your Executive team?

Have you chosen any form of AI Framework? We use the APEX Framework

The most straightforward way to drive productivity is to identify what is holding your people back.

Look at your organisation from End-to-End. Avoid focusing on “issues” without an holistic view of relationships and dependencies.

Every business is a complex web of interconnected relationships that will include people, processes, systems, and risks. Driving change without a clear view of the implications is a recipe for failure.

70% of digital transformations traditionally fail.

It is looking like this failure rate is higher in AI specific transformations where MIT (2025) report 95% of AI Pilots fail to scale to business solutions and Gartner (2025) state that 85% of AI Projects fail to deliver business value!

The reality has not changed. It is rare for the technology to be actually faulty. The issue usually lies in the work before that point; the how and why, not the solution selected.

If you aim to become AI-ready, foster a data-driven culture, and enhance productivity, consider taking the following steps to initiate the process.

• Business Process Model or DTO (Digital Twin of your Organisation): defines how your business actually works and includes a full relationship model of your organisation, including your People, Processes, Systems and Technologies as well as Risks, Controls and Mitigation and a full RACI matrix for every object, process and risk across the Business Landscape.

How do you start a DTO?

• Business Glossary; you cannot be data-driven and AI-Ready without the discipline of a Glossary which defines every metric, goal, and KPI, down to the data source. Ideally linked to your Data Catalogue.

See more here – How to start a Business Glossary

• Business Strategy: everything is tracked back to the business strategy ensuring every “tool” is delivering value against one or more strategic objectives.
• AI-Framework: Using an AI Framework whereby all AI initiatives can be maintained within a Cost, Benefit and Risk framework.
• Business Leadership: AI implementation with be less effective if it is not controlled within a Business needs led lens, which should be the first part of your Framework. We use the APEX Framework.