We’re very interested in this too and believe AI has huge potential. But before any deep or automatic integration, we want to take one important step first: a safe, fully controlled experiment - and we invite you to try it with us. Your feedback will directly shape MLO’s future in terms of AI.
Caution matters
Your MLO database contains years of thoughts, plans, and commitments. While modern LLMs are powerful, they still make mistakes - and your data deserves protection.That’s why any AI use in MLO must guarantee:
Full transparency - nothing happens behind your back
User control - you approve every change
Easy verification - clear before/after comparison
Full rollback - nothing is irreversible
A simple experiment you can try now
Here’s the idea: instead of letting an AI touch your live data, we use MLO’s existing export/import capabilities and keep you in control at every step.
Step 1: Export your data
On desktop, export your entire MLO file to XML. This format already contains everything an AI needs: hierarchy, task properties, dependencies, goals, projects, and flags. And it’s all plain text.
Step 2: Share the XML with an LLM
Pass the XML to an LLM (for example, ChatGPT) and briefly explain any default values if needed.
Step 3: Ask a focused, well‑defined request
For example:
- Find the 'Build a house' project and merge duplicate tasks
- Move this project to My Goals folder
- Analyze related projects and mark the most important ones with a red flag
- Find inbox tasks from the last 5 days related to this project and move them there
Important rules for the AI:
- Return the result in the same XML format
- Do not change anything unrelated to the request
- Provide a detailed list of all changes, with explanations
Step 4: Review the result
Now you have:
- the original XML
- the modified XML
- a clear explanation of what was changed and why
You can compare them and decide whether the result makes sense.
Step 5: Test safely
If you like the result, import the modified XML into a new MLO file. This keeps your original data completely untouched.
Yes, this means setting up sync again (connect it to a new Cloud file) - but it allows you to test the entire concept with zero risk.
Why your participation matters
This simple real-world testing can answer the key questions:
Does AI actually help with real MLO data?
Does it save time on inboxes and large projects?
Is it valuable enough for deeper integration?
If this approach proves helpful and reliable, then creating a more convenient, built‑in mechanism becomes a realistic next step - without changing the core data model and without sacrificing control.
Please share your results with us at support@mylifeorganized.netor in the Community group. Even short feedback is extremely valuable. Thank you!
The MyLifeOrganized team
0 Comments: