Built on your accounts, in your name.
The worry with putting AI into a business is a system that leaks data, or acts when nobody asked it to. Here is what we build against that, and what we do not claim.
Your accounts, your data
Nothing we build is a platform of ours that you rent.
It shows you before it acts
Our own assistant works this way:
- Before it sends an email, a DM or a post through its connected tools, its owner is told where it is going, shown how it starts, and given time to stop it.
- By default he has three minutes to stop the first send of a task, and 45 seconds for each one after it.
- Its sends are capped each day.
- If he cannot be told, nothing is sent.
We build yours on the same pattern.
When a send counts
The free report we built for ELM Tree Marketing, our sister company, is built to email itself, and a send only counts once the email is found in the sent folder.
This site, as an example
- Only this site's own scripts can run on its pages.
- No other site can show these pages inside its own.
- The form sends your message only to us, and each internet address can send it at most five times in a ten-minute window.
- The site sets no cookies and runs no analytics or advertising scripts.
- Security problems can be reported to ethan@elmtsystems.com.
What we do not claim
We do not claim any system is impossible to break. We do not claim an outside security audit. The announcement covers emails, DMs and posts. We do not claim that everything our assistant does through a connected tool is announced first.
How it starts
Start with the audit, or go straight to a build if you already know what you want. You can stop after any one of them.