Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Muse – Meta’s personal AI agent

ai.meta.com

Post reply on HN
Print view View original

371–380 of 723 posts

Re: Muse – Meta’s personal AI agent

#371

Post by bdangubic »

smokel wrote: post #120
1123581321 wrote: post #109

Earlier quoted context omitted.

Yes, of course. Wikipedia has a quick summary of the most popular networks ranked by monthly active users. Facebook is #1 and Meta owns three of the top five. https://en.wikipedia.org/wiki/List_of_most_popular_social_pl...

Those are self-reported numbers from 2023. Besides most likely being highly inflated, I think Facebook is on the decline.

facebook is everything except on the decline, HN bubble is the only place anywhere where this is being thrown around without any rhyme/reason

Re: Muse – Meta’s personal AI agent

#373

Post by russellbeattie »

nickysielicki wrote: post #351

People are unfortunately just ranting about not trusting meta, which… okay, fine, but having played around with this for the past hour or so it’s probably my favorite mobile LLM experience so far across Claude and ChatGPT and Gemini. It’s very good and in its own class.

Agreed. I'm not a fan of Meta in any way, but I've had fun messing with the app this afternoon - both on my phone and the web version at muse.ai - to see what it could do.

Like the announcement said, it's running in its own VM, but it's an impressively transparent system: You have access to its System Files right in the GUI, which surprised me. You can look at the AGENTS.md and other harness files it uses as well as some shell scripts, bash files, cronjob folder etc. It seems a lot more obvious to me how Muse works than say Claude or ChatGPT, which don't make it nearly as clear where stuff lives and how long its there for. I mean, I get they use VMs as well, but there's definitely no "System Files" menu option in those phone apps at least. The Muse bot seems much more "permanent" in my mind as a result.

The AI is quite helpful as well. It let me download its 350Mb "hatch" binary which runs the bot/harness/agent (written in Rust), plus it zipped up the library of skills it uses and all the system files in my VM for me to look at on my machine. Nothing extra special, but it was nice to browse through all the files and the binary strings to see how it ticked. It also explained the limits of the VM to me - for example, that I can't SSH into the VM as it runs through a proxy and can't be reached directly, but was willing to help me add itself to my Tailscale network.

It's basically a nice looking UI on top of a typical agent harness using standard Markdown files, all running in a little Linux cloud VM spun up just for you. And Muse Spark 1.3 is finally good enough that it was actually useful (and fast), so I can see why they launched the app/service now.

From a non-geek perspective, I asked it about selling my old Nintendo Switch on Facebook Marketplace - something I've been meaning to get around to for about a year or so now - and it said I can upload a picture of all the stuff I have and it'll check prices and create multiple listings for me. That alone might make me hate Meta a little less after this if it works as advertised.

Re: Muse – Meta’s personal AI agent

#375

Post by zrobotics »

enahs-sf wrote: post #359

I just asked it: “Can you lowball fools on marketplace for me?” It replied: “Say less, I can do that. What am I hunting…” People are going to use this product for bad things.

Oh god, as if marketplace wasn't already enough of a toxic cesspool. I miss Craigslist more and more every year, unfortunately FB marketplace seems to be where everyone in my area has congregated.

Re: Muse – Meta’s personal AI agent

#378

Post by welder »

ElliotPadfield wrote: post #321
tonyhart7 wrote: post #235

Earlier quoted context omitted.

"This account needs permission to view this page The account you're signed into doesn't have access to this page. Please log out and follow the steps in your email." there is no logout or instruction whatsoever on email, so zero tech support definitely not true

Also running into this same issue. At first I logged in fine but nothing in the UI would hydrate properly and now just hit with this error.

Same here. Facebook has a horrible experience when you're traveling or logging in from a new computer. I still have business Instagram accounts blocked with no way to login and no support. Wouldn't want to depend on anything from Facebook, especially not something with access to my email, calendar, and browser.

Re: Muse – Meta’s personal AI agent

#379

Post by hokumguru »

jmathai wrote: post #340

> Your personal data is safe and secure That's from the article. I can't think of anyone I'd trust less with my data than Meta. Well, thieves, I guess.

I trust Meta far more than, say, Anthropic. For one, the judgments against them from the government just mean that they have tons of oversight and auditing happening now. The other providers are just the wild west with privacy.

Re: Muse – Meta’s personal AI agent

#380

Post by jddkj »

asdfman123 wrote: post #319
Grombobulous wrote: post #294

I think the product as shown in this marketing material with these examples falls squarely in the “nobody asked for this” territory of AI. At least with B2B AI there is a problem to be solved: reduce my labor costs and increase my business output. What’s happened here is that consumer technology is already solved, but companies like Meta are trying to jam AI technology into something that solves the non-problems that…

No, after seeing how powerful AI is at my job, I definitely want AI to help plan dates, book flights, generate spreadsheets, do all the things. The only problem is, like someone else pointed out, that I don't trust Meta.

This is where I'm at as well. I've been using AI increasingly more and more at work and it has been genuinely useful. I want a Siri-like assistant that is actually useful. Muse seems like it's that tool. Except, it's Meta.
Post reply on HN