Applying AI without a rigorous pre-implementation assessment is dangerous. It can cause more damage than a bull in a china shop.
Undertaking a systematic, enterprise level pre-AI Implementation assessment is essential.
First assess the current state, is the data:
- Managed effectively?
- In multiple data silos, or bubbles?
- Accurate, authentic, and authoritative?
- Well governed?
Then, moving on:
- Create a roadmap, possibly a consolidating data from a range of systems.
- Apply consistent metadata.
Now you can formulate and apply an AI strategy that maximizes:
- Workflow.
- Security.
- Return on investment (ROI).
This pre-AI assessment workshop explores how to address the fundamental questions:
- What are your AI goals?
- What data do you have to train the AI?
- Where is your data?
- How will you access and distribute your data?
- In what ways are you allowed to use your data?
- Is all the data really yours?
- What technology will you use for AI?
- How will you govern and evaluate your AI?
Good, trusted, authentic data is critical for implementing AI. The certainty that the data is accurate and permissible to use is essential for success.
And don’t forget that people remain people – be mindful of the people, processes, and technologies that may influence the data and learning within the business.
