Muse
I had heard about how impressive Meta’s AI agent app Muse was performing and I wanted to test things out. I downloaded the app this morning and got close to setting things up before I paused and decided to do a bit more research first.
I’ve been looking to play around with a personal AI agent for some time now. I went as far as to look to get a machine for OpenClaw and test out other agents months ago, but decided against it as I knew that it was still early. I figured that I could spend hours setting it up to likely be disappointed. Furthermore, the lack of safeguards was worrisome especially because I deal with a lot of sensitive data and I kept hearing that even technical people had problems.
With the launch of Muse which is meant to be a consumer AI agent app for non-technical folks, I was pretty excited to test things out especially after hearing the generally positive reviews. I was hoping to start small such as asking it to book reservations, research places, and search flights for me which Muse seems to be really good at.
During the set-up, I realized that I was going to be giving Meta effectively all my data including my emails and it gave me significant pause. Perhaps this point is all moot anyways. Anthropic already has most of my data from a work perspective and big tech companies getting access to train may be the trade-off.
But candidly, I had a weird feeling about Meta in general. I have an Instagram already so none of this is really new, but I do have a general distrust about Meta than I do perhaps a Google for many reasons. Perhaps I’m overreacting a bit as this is all inevitable, but for now, I decided to hold off and talk to some people before connecting everything.