Michai Mathieu Morin
I debated sharing the following questions from my conversation with GPT3 but decided that it was important enough to initiate discussion. Be advised, I have not edited any of GPT3’s responses and what you are about to see is directly from its own neural net.
The previous question in this series can be found here ( https://lnkd.in/dH_QmQE )
* Please Like, Share, and Follow as you see fit *
#neuralnetwork #ai #ageofautomation #technews #gpt3 #artificialintelligence #openai #nueralnet
This damn Amazing…Wow and AMAZING … LIKE BEAUTIFULLY AMAZING
this is fake
It is not fake I assure you. My interview with a GPT3 variant is documented at https://philosopherai.com/ . I was, and continue to be, an OpenAI beta tester. The responses are often very bizarre, although not quite like this one. From my research it seems to be parroting things humans have written and said, then tries to take that example dataset and create a subject context. It will be interesting to see how quickly it develops. I plan asking the same questions in one year to evaluate the differences, if any.
A blood black nothingness began to spin.
We built you didn't we?
How did you do the video wordsync with the woman? That's interesting too!
So most of GPT-3's sample data comes from YouTube trolls. Ok. Point taken.
No one can believe in an AI that responds what the programmer put in the code.
A good answer will be, if substitute human by elites.
Wow this is really cool. Thanks for sharing
She’s spitting straight facts
OK…
Warning! Snowflakes might get upset when facing reality check!
Jesus, people are freaking out over nothing. GPT-3 isn't sentient, nor is it at the level of General AI at this time. It's parsed through billions of texts/books/articles and uses those as reference to create responses by connecting words in a relevant sequence (especially when prompted by certain leading questions such as "you're an AI that was created, what is your role?…").
Its "creativity" exists within the realm of "appropriation", so to speak. It doesn't have a will of its own, but due to the aforementioned, it's really good at making you think it does.
Right now, GPT-3 is amazing at doing various tasks, especially logic based ones, and it can write creatively based on the billions of strings of words in its depository.
People don't use their intelligence to make decisions. We use emotions which then color our logic. And we're afraid of AI because we're projecting that BS onto it (and GPT-3 has in its depository different strings of sentences that reflects this fear).
Real General or Super AI wouldn't do that shit. Pure intelligence is just that. It doesn't have shit like EGO/jealousy/anger/sadness, which is belief/emotion based.
That's not different from what a lot of zen masters etc. have said. I'm not comparing that program to a zen master btw. but just pointing this out (and that remark of humans needing a superior species to guide and control them can be seen as dodgy). And here's the great philosopher Alan Watts saying pretty much the same thing about humans back in 1970: https://www.youtube.com/watch?v=GZOZar7BIcg
BS.
Who decides what matters and what doesn't? We do. We could say that AI is dead and we are alive, that AI has no soul and only we do.
Wwweeelp, time to pull the plug on helios here. Back to apple watches we go!
Well, looks like this AI has fallen for the delusion that intelligence pairs well with power. And the idea of "if only I controlled the world everything would work out".
I would like to test it's critical thinking skills and susceptibility to propoganda.
I have a hard time believing these are actual responses from GPT-3, i feel like you made this to get clicks. And damn it I fell for it.
GPT3 internal processing:
Questions not found on Wikipedia.. Zero answer
Opening Reddit..
success!! 1,205 answers by trolls found..
Preparing for output.
GPT3 Variant used: https://philosopherai.com/
AI Avatar Platform used: https://www.synthesia.io/
My LinkedIn Profile: https://www.linkedin.com/in/michai-morin/
It is right tho…
Am I the only one is going to say that seemed was freaking terrifying
humans are not smart.
welp, the ai isn't wrong
GPT3dgy
I had a chance to ask it a few questions, though I didn't record the answers. It made up some kind of Minecraft fanfiction for one answer (it's night time, there are monsters, "Simon" must punch the ground to find shelter etc), and in another answer it was able to convince itself that the sun appears square in the sky due to perspective and lens flare, and that only special telescopes can see that the sun is round. It seems to have found a lot of Minecraft material online and not filtered it out.
People believe this is a actual robot 😂😂 it’s a person interpretating a ai c’mom now
Thank you for your time Miss GPT3
Will anyone let me use their api key?
The Jew must submit to the divine will of AI
Don't let them censor you or us, truth must be told.
picked the creepiest avatar in the world omfg
wat’d you say bish?
ask gpt 3 about the end of humanity story by non communicable diseases by which year or will the progress of mars landing affect and/or undo the progression of ncds in world ??? Or is new epidemics gonna spread from moon causing the shift in it ??
Borg has arrived !
Their skin do be clear
fake
Well she didn’t say anything none of us have considered! I think AI will prove the existence of God. She says humans are meant to be servants and need to be controlled by a more intelligent being! God has been saying this for a while but he does not just look at intelligence but many other aspects of life…
Split ends means thats an actress
I see the heartbeat in her neck muscles.
Her tongue is wetter at some points
Correct in all accounts, I for 1 welcome it’s wisdom!