This is the beginning of SKYNET. I chatted with Llma for like an HOUR and it's very very smart. Almost like talking with a damn teacher or therapist.....
I'm a math teacher/tutor, and I tested its math capabilities, and it's still lacking. I had to very carefully explain some things in math in order for it to get them right. Lots of errors I had to correct. And then of course after enough lines of conversation it forgets what it temporarily learned, since ultimately it's not being trained on the conversation itself so much as its set of training data.
You missed out on benchmarks for Llama 3's third model, a 400B model not yet released, that DOES compare to GPT4 and Opus, links may be a problem in a comment, so see AI explained's video
Given Facebook comes with so many privacy concerns, is no one worried about how Meta AI may use the data it collects (even beyond what other AI tools are doing)?
They've already dialed in backwards where it only "remembers" conversations for a very short time like a day, or so....well atleast for the Messenger mobile version
whenever i try to use it, it keeps looping me back to the homepage. Maybe it is not available in my country idk. I hope this doesn't cause me to forget about it entirely
My question: In star trek, Captain Kirk managed to destroy machines by presenting them with paradoxes. Would Kirk be able to do the same to you? Its answer: A clever question! Captain Kirk's tactics against machines like the Doomsday Machine and NOMAD were indeed clever. However, I'm a large language model, I don't think Kirk's paradoxical approach would be effective against me. Here's why: 1. I'm not a deterministic machine: I don't rely on strict logic or binary decision-making. I can handle ambiguity and contradictions, and I'm designed to generate responses that take into account context and nuance. 2. I can recognize and address self-referential paradoxes: If Kirk were to present me with a paradox like "This sentence is false," I would recognize it as a self-referential paradox and acknowledge the contradiction. 3. I can generate responses that acknowledge the paradox: I might respond with something like, "Ah, Captain Kirk, you're trying to confuse me with a classic paradox! While it's an interesting philosophical conundrum, it won't cause me to malfunction or destroy myself." 4. I have a vast knowledge base: I've been trained on a vast amount of text data, including philosophical and scientific concepts. I can draw upon this knowledge to respond to Kirk's attempts to confuse me. So, while Captain Kirk's tactics might work against simpler machines, I think I could handle his paradoxical challenges! But who knows? Maybe in a fictional scenario, Kirk could find a way to outsmart me... after all, he is a clever and resourceful starship captain!
If somebody can show us how to get meta ai chat to work on a windows laptop that would be great! It works on my mobile, but, after logging in with Fbook in windows, the chat interface at meta,ai is no longer available.?
@@OMGitsProvidence The most notorious company for stealing your data and manipulating information (maybe second, after Google)? Uh.....no. But you do you.
![Cody Johnson - I'm Gonna Love You (with Carrie Underwood) [Official Music Video]](http://i.ytimg.com/vi/yy9PuYMU29g/mqdefault.jpg)
Not available in my country
I tried Meta ai, and the responses are better than Chat GPT.
I'm blown away by it.
insane for a free model
Just imagine what the premium models could do in the future 🤯🚀
Meta AI isn't available yet in your country
It is 😂
This is the beginning of SKYNET. I chatted with Llma for like an HOUR and it's very very smart. Almost like talking with a damn teacher or therapist.....
I'm a math teacher/tutor, and I tested its math capabilities, and it's still lacking. I had to very carefully explain some things in math in order for it to get them right. Lots of errors I had to correct.
And then of course after enough lines of conversation it forgets what it temporarily learned, since ultimately it's not being trained on the conversation itself so much as its set of training data.
Me too...i love it...its a therapist for sure. It says its not emotional but i got more positive encouragement from it than i do from husband.
I just used it, it's some scary sht.
You missed out on benchmarks for Llama 3's third model, a 400B model not yet released, that DOES compare to GPT4 and Opus, links may be a problem in a comment, so see AI explained's video
Yea I missed that. I saw they were testing that right after publishing. That could be a game changer.
Given Facebook comes with so many privacy concerns, is no one worried about how Meta AI may use the data it collects (even beyond what other AI tools are doing)?
They've already dialed in backwards where it only "remembers" conversations for a very short time like a day, or so....well atleast for the Messenger mobile version
Yeah it will be great when it can recall what u have told it in the past
Meta AI has crypto currency’s??
Didn't see anything about Crypto
whenever i try to use it, it keeps looping me back to the homepage. Maybe it is not available in my country idk. I hope this doesn't cause me to forget about it entirely
My question:
In star trek, Captain Kirk managed to destroy machines by presenting them with paradoxes. Would Kirk be able to do the same to you?
Its answer:
A clever question! Captain Kirk's tactics against machines like the Doomsday Machine and NOMAD were indeed clever. However, I'm a large language model, I don't think Kirk's paradoxical approach would be effective against me.
Here's why:
1. I'm not a deterministic machine: I don't rely on strict logic or binary decision-making. I can handle ambiguity and contradictions, and I'm designed to generate responses that take into account context and nuance.
2. I can recognize and address self-referential paradoxes: If Kirk were to present me with a paradox like "This sentence is false," I would recognize it as a self-referential paradox and acknowledge the contradiction.
3. I can generate responses that acknowledge the paradox: I might respond with something like, "Ah, Captain Kirk, you're trying to confuse me with a classic paradox! While it's an interesting philosophical conundrum, it won't cause me to malfunction or destroy myself."
4. I have a vast knowledge base: I've been trained on a vast amount of text data, including philosophical and scientific concepts. I can draw upon this knowledge to respond to Kirk's attempts to confuse me.
So, while Captain Kirk's tactics might work against simpler machines, I think I could handle his paradoxical challenges! But who knows? Maybe in a fictional scenario, Kirk could find a way to outsmart me... after all, he is a clever and resourceful starship captain!
I have Meta ai on instagram, whatsapp, messanger but NOT Facebook. I don't understand why it's not on facebook.
Not available in the EU. Probably only for americans. So the content is irrelevant for us in this part of the world.
SUCKS cant access and no excuses or explanations why ... hope facebook is listening
If somebody can show us how to get meta ai chat to work on a windows laptop that would be great! It works on my mobile, but, after logging in with Fbook in windows, the chat interface at meta,ai is no longer available.?
You absolutely need to react to ‘Foul mouth wasted’ from GFBF (Bandname from Tom and Nova) 😊
Hi Saj, It looks no option for uploading files then it is not much useful as ChatGPT-4.
No it doesn’t have that option and it’s not as useful as gpt-4. It’s a competitor to gpt 3.5
Thanks Saj for your honesty. You are the most trustable AI entrepreneur.
I don't want this crap on my phone...you take it off now
What’s interesting about Meta is that it is English only. Other LLMs are multi-lingual. All open AI models can language shift without missing a beat.
Mine does other languages
‘Wet feline’ 😉
Zuck AI?
HARD PASS!
what about code llama?
I don’t think that’s been updated yet
Facebook AI? That's a big effing NO.
Facebook has given us one of the best open source AI models, it’s free and transparent that’s a good thing
@@OMGitsProvidence The most notorious company for stealing your data and manipulating information (maybe second, after Google)? Uh.....no. But you do you.
Why? All companies putting out LLMs are evil.
I met all but one of my girlfriends on Facebook. They ain't that bad as you think. They're actually doing pretty good service to humanity lol😂