Garbage in, Garbage out: Applies to AI too
Article 2 of 3 in a short series about PEP and AI
When AI Meets “Final Version”
A few years ago, many of us spent a great deal of time searching for documents. Today, AI can often find them in just a few seconds.
That sounds fantastic. And it is. At least until AI finds:
- Budget_final_v2
- Budget_final_v2_new
- Budget_FINAL
- Budget_FINAL_latest
And then asks the only reasonable question:
Which version do you mean?
It is both slightly amusing and somewhat revealing. Because while AI is impressive, it is still dependent on the material we give it access to.
Garbage In, Garbage Out
The expression may not be particularly elegant, but it is difficult to find a more accurate description.
If the foundation is unclear, the result will be unclear too. That applies to people, and it applies to AI.
When information is scattered across emails, Teams messages, OneNote notes, desktops, and people’s memories, it becomes difficult to create a complete and reliable picture. Not because AI is inadequate, but because the foundation is unclear.
Many people hope that AI can bring order to chaos. To some extent, it can. But if you struggle to find the correct document version or remember where a task was noted down, or why it was not, AI will face a similar challenge.
The technology itself is not necessarily the problem. More often, the underlying data is incomplete.
Clarity Before Automation
When we work with the PEP principles, we focus on creating a clear overview of tasks, keeping calendars up to date, and establishing effective work habits. Does that become less important when AI enters the picture? Quite the opposite.
AI works from the information available to it. When data and workflows are clear, AI can be a valuable assistant. When they are unclear, its recommendations become equally uncertain.
Four areas are particularly important:
- Tasks: Can they be found in one place, or are they scattered across different systems?
- Calendar: Does it reflect the time you actually have available?
- Documents: Is the current, final version easy to identify?
- Commitments: Are decisions and planned follow-ups clearly recorded?
The structure does not have to be perfect. It simply needs to be good enough for both people and AI to understand how things fit together.
The Calendar Never Lies. Or Does It?
The calendar is a good example.
If it contains meetings, focus time, preparation, follow-up activities, and time away from work, it can provide a reasonably realistic picture of your capacity.
If the calendar only shows meetings, both you and AI are working from an incomplete picture of reality. It is a bit like putting together a jigsaw puzzle with several missing pieces. You may get an idea of the final image, but you never see the full picture.
The same applies to your task list. If some of your important tasks exist only in your head, they are invisible to everyone else. Including AI, which, despite its capabilities, is not a mind reader.
That is why an up-to-date calendar and a consolidated task overview do more than create order. They improve the basis for the decisions that both you and AI make.
A Quick PEP Check
The people who get the greatest value from AI are not necessarily those who write the most sophisticated prompts.
More often, they are the people who know where their information is stored, keep their tasks organized, use their calendars actively, and follow up on commitments.
In short, they have created a working environment that both they and AI can navigate. As a result, they receive better support.
Try asking yourself:
If AI were helping me prioritize my next week, would it have access to a complete and reliable picture of my work?
If the answer is no, try to identify the underlying cause and start there. It could be your tasks, your calendar, your documents, or your inbox.
Today’s Key Message: The better foundation you create for yourself, the greater your ability to use AI as a source of qualified support.
In the next article, we will take a closer look at where AI can provide the greatest value in everyday work and share a few practical examples of how you can put AI to use.
Our job is to make you better at yours!