Here at Hart IT we aim to take a pragmatic and realistic approach to “AI” (or predictive text generators as they should be called!).
There are real risks with any IT project today that an unscrupulous or unwise creator could take shortcuts and use AI instead of doing the job properly. The end result is a vibe-coded project, that will meet the minimum requirements but be an unmaintainable, unscalable, & insecure mess.
At Hart IT we aim to build dashboards that are as reliable as possible, and if needed, as easy to fix as possible. To achieve this we have developed our own processes and standards which we are continually improving on. This means the dashboards we build for our customers can be changed or scaled efficiently and consistently.
To ensure security, we are always making sure that our customers’ data is never transferred to tools or systems outside of our direct control or that are insecure.
We will use AI as a tool when it is useful and can save us time, but we are consciously not following the prevailing trend of slapping an AI label on everything and charging more for it.
Our AI Policy at Hart IT is:
- Don’t use AI for the sake of it.
- Don’t avoid using AI because you dislike it.
- Always check or test anything produced by an AI tool
- Never put any client information or data into an AI tool.
- Never give an AI tool access to any company files, data, or software.
- If a client can tell AI has been used, you are using it wrong.
- Don’t use AI to do anything you don’t have an understanding of.
- When using AI for code, don’t “black box” anything.
This policy will be refined and altered as technology changes.