cross-posted from: https://lemmy.world/post/52306182
"Behind the scenes, Meta is having some of these calls being handled by humans in call centers, 404 Media reports, despite employees practically begging the company not to go through with the charade. "
It would be an impressive example of a practical AI use case for the average person — if it were really, truly, AI. Behind the scenes, Meta is having some of these calls being handled by humans in call centers, 404 Media reports, despite employees practically begging the company not to go through with the charade.
A(ctual) I(ndians) strikes again
I feel like as soon as AI Agents calling businesses becomes common, a lot of people will simply hang up on them when they call.
Nah - by then it’ll be AI agents answering too…
Spending millions of tokens on text to voice to text for a meeting that could have been an email
They’re going to figure it out and go back to sounding like modems, and saves the spare tokens to plot our demise.
Who is the they there? The chats bots would need to be programmed to do this, they won’t figure it out on their own.
They are running through a predefined logic tree for any action that isn’t next token prediction.
They need strict guardrails. It’s why it’s bullshit when they claim “it” hacked some external system. Someone programmed it to do those things, the chat bots didn’t evolve the ability. These companies play the media and consumers for fools.
For text, they kind of already are.
Each agent was given a name, personality traits, a role such as conflict mediator or resource strategist, and an overarching goal for its “life” in a shared town… By the end, each simulated world had undergone its own shift in language. The world inhabited by copies of OpenAI’s GPT-5.5 model, for example, compressed speech, removing grammar and shortening messages. “Friendly voltage counted. If the handle needs your mouth, it dies anyway,” one message read.
That description is nonsense though, they weren’t running multiple copies of gpt, it was a bunch of programs maintaining separate context using the same LLM back end. The LLM was just talking to itself in a closed loop generating plausible text each time it got looped back on itself. Saying it got compressed is a generous way of saying it devolved to nonsense.
Someone’s agents will eventually market us agents to deal with the other guy’s agents until it’s agents all the way down.
They’re currently marketing haggling bots to both sellers and buyers for auction sites, so that two bots owned by the same company can argue each other back and forth trying to squeeze the price down. Truly brilliant.
This is unironically the end goal as ut means billing multiple people for every interaction.
It won’t be economically feasible, but it is the goal.
Good thing Muse hands things off to humans in a call center then, huh? /s
its a feature!
That’s a downvoteable title
Not when it’s pointing out another failure by Zuckyboy
The Turks are at it again.
Is there a client or way of posting that will consolidate 4 shares to 4 communities I’ve joined into one post? Using Boost/Android currently.
I use interstellar and when I view a post that’s been cross posted it shows all comments and replies from all cross posts regardless of which crosspost I click on.
Is that what you’re looking for?
Nice! The UI doesn’t feel as polished as Boost, but that’s a great feature.
