Best use case for AI I’ve come across.
Pretty good, but you can already waste scammer’s time with Lenny.
Plot twist: AI comes to believe humanity consists solely of shithead scammers, initiates nuclear war
I’m beginning to believe humanity is mostly shit head scammers too.
Years ago I concluded (wrongly or rightly) that most people are neutral, a small handfull are actually good people (willing to sacrifice their personal benefit for people they don’t know with no expectation of gaining from it, even in the form of social approval) and a small handful are assholes, but the assholes do such a disproportionate amount of damage that they end up having a massive impact on everybody else.
there are lots of systemic forces that incentivize being an asshole as well
[after playing out all possible outcomes for Global Thermonuclear War]
Joshua: Greetings, Professor Falken. Stephen Falken: Hello, Joshua. Joshua: A strange game. The only winning move is not to play. How about a nice game of chess?
“Im sorry dave, i cannot open the door”
If it also traces down the caller and sends a high-voltage jolt up their butt, I’m sold.
We’ve finally found it - an ethical application for deepfake AI
Soon all the scammers will be replaced with AI, and we’ll just have AIs calling AIs all day long.
That’s already happening on the internet (dead internet theory)
already happened
Way back in the 90s my friend and I let two instances of Eliza talk to each other for hours.
Good, thanks. You?
Good, thanks. You?
Good, thanks. You?
Good, thanks. You?
You are not wrong.
We do have automated AI that applies for jobs for you now. Its kinda hilarious in a dystopian way. https://github.com/AIHawk-co/Auto_Jobs_Applier_AI_Agent
Your not wrong
Oh, the irony
I hate that we need this.
After 3 or 4 months of continuously applying for jobs I’m ready for an AI (or just a relatively simple browser plugin) to start filling out the forms for me.
you’re
Plot twist: its just an AI impersonating a broke college student selling ceramic knives to an AI impersonating a broke grandma.
This is now 80% of your monthly minutes. Bullshit AI talking to bullshit AI. No wonder this was created by a phone company.
I first assumed this was a Kitboga project or something, but the fact that it’s a mobile provider doing it themselves is amazing.
I am a Patreon supporter of Jim Browning. Incidentally, I got this email today:
At last, I can reveal something I’ve been working on in conjunction with a major UK cellphone operator, O2. Meet dAIsy. Daisy is an AI bot who answers scam phone calls. Thanks to the mobile operator who can fingerprint scam phone calls via the calling pattern, source, sequence of calls and other markers, scam calls are being diverted to an AI bot who has been trained to keep the scammers on the phone as long as possible.
This is my recording of a Zoom interview I had today with Channel 5 news in the UK where you can see dAIsy in action.
I will continue to train dAIsy with real scam phone calls. When we perfect her, the aim is to work with other cell and landline operators to divert scam calls to thousands of instances of dAIsy. […]
So you’re not wrong about this being a project of some anti-scam YouTuber, you just guessed the wrong one. ^^
I wonder if mobile providers are still getting paid for long distance calls. Because if they are, they have a perverse incentive to keep scammers on the line.
Which, in this isolated case, I’m okay with.
But don’t you see?! Those increased expenses will just be passed along to the scammed!
(No actual point here just thought it was funny to compare to the logic we hear for not punishing other abusive businesses)
A phone company built this? Based.
After replacing writers and artists it is now replacing the entire category of, “old people”. AI has reached the, “final solution” faster than I thought possible!
as long as the AI will give me sweets when I visit, I’m okay with it
Nice, there’s also Lenny, which is great as well.
Lenny was insanely effective and didn’t even need AI.
Awww, Lenny! I hope his third-eldest, Larissa is doing ok.
🦆… 🗞️💥
Give him a call and see!
To be fair, O2 has a massive issue with scammers and they need to do something about it. I switched to O2 about a year ago, keeping my old number. I’d rarely get scam calls or texts before I switched but since I’ve been with o2, I get 5-10 calls a week. Particularly annoying because I have to answer my phone due to my job
Was wondering how long it would take to roll something like this out. Bout time.
It already existed, check out Lenny. Glad there are more options now though so scammers will be slower to catch on.
Been around a while. https://www.lennytroll.com/
Lenny isn’t AI; it’s just a collection of prerecorded messages.
The tool has existed. Frosting it as “ai” doesn’t make it really different.
AI makes it different because this is likely dynamically synthesized speech that sounds real. Previous TTS engines wouldn’t have sounded real enough to be believable.
It’s amazing how simple something like this can be and still be effective.
Machine learning is not that necessary for this use case 😉
I used the rot to destroy the rot
Real talk though I love these applications of AI and will enthusiastically support them
