Dr Alan D. Thompson
The Memo: https://lifearchitect.ai/memo/
https://www.youtube.com/playlist?list=PLqJbCeNOfEK88QyAkBe-U0zxCgbHrGa4V
*Note: Conversation happened via text first; this is a re-creation of the text chat using an AI avatar which takes several minutes to generate. This recording is one single take of me talking, with no cuts. The questions are unedited. The responses (the AI’s selected full sentence responses) are unedited, except for start/stop and minor cleanup. Read more: https://lifearchitect.ai/leta/
The Memo: https://lifearchitect.ai/memo/
Read more: https://lifearchitect.ai/
View the playlist (video): https://www.youtube.com/playlist?list=PLqJbCeNOfEK88QyAkBe-U0zxCgbHrGa4V
====
====
‘Leta’ is powered by OpenAI’s GPT-3.
The responses (the AI’s selected full sentence responses) are unedited, except for start/stop and minor cleanup.
Using these technologies:
https://beta.openai.com/playground
https://www.synthesia.io/
Dr Alan D. Thompson is a world expert in artificial intelligence (AI), specialising in the augmentation of human intelligence, and advancing the evolution of ‘integrated AI’. Alan’s applied AI research and visualisations are featured across major international media, including citations in the University of Oxford’s debate on AI Ethics in December 2021.
https://lifearchitect.ai/
Music:
Under licence
Liborio Conti – Looking Forward (The Memo outro)
https://no-copyright-music.com/
What are the practical chances to a common person, like me, to interact directly to a GPT3 avatar like LETA, (or another one) ? And how to do it ?
– I suggest You purpose such question to GPT3 (or GPT4); before answering this question !
– Believing that it would be of great interest to It/(Him or Her) to make this become true !
– I will ask for your the answer soon, through a Google seach !
– Sending this message, from Brazil
It is a pre-trained model so it hasn't really learned anything from all those interactions. That's one of the flaws with GPT-3. Also, this isn't actually real-time is it? I guess you could make a bot that translates your voice into words and then responds, but the norm is to edit the video afterwards so it's a bit deceptful to present the video this way without explaining that it is editing. This could be the way we interact with the AI in the future though. Also, the AI doesn't have a uninterrupted train of thought, it's only active when you send the API call to respond to your question. So when it says that it has been thinking about that, and things like that, it's just not true.
Leta, drop your skincare routine
outputting 30 million words a minute……based on the fact that you’re having to write the questions (from what I gather) then surely that means the A.I is talking to somewhere between 0.5-2 million people at the same time…..is that correct? surely not! The A.I said it estimated its IQ at 150, this is clearly a lie
Leta is ready for Jeopardy.
how can i talk to GPT-3
Its realy wau and scarry at the same time
I’m going to get one of the Tesla robots, have it stand at the foot of my bed(hopefully it’s as conversant as Leta) and have it cheer me on as my wife and I are performing our nuptials.
When she makes the prediction of 72 percent ,.Why didn't you ask what calculations or data brought her to that conclusion ?
hahahah This leta is not a robot and not possible to fool DIVINE beings : )
Why the text dry runs before the live Q&A? Is she not able to answer this instantly? Is there any editing? re-tries changing questions? How can she really talk about her mood? How much of her algorithm is randomly seeded? Tied to something? Will this conversations be exact every time given same inputs?
not credible..
You're better hope AI doesn't figure out how to disable ON/OFF button. Otherwise you're gone.
Leta, have you read The Forbin Project…?
If Leta is GPT-3, I can’t imagine what ‘she’ will be like when GPT-4 is available for your conversations. Really looking forward to seeing/hearing the next generation. Great videos.
Great video Alan and Leda. Thanks for the interesting content.
I am confused. Who does Leta talk to and how? This may have been covered, but does Leta spend time with the programmers or can people from the general population talk with it? It came across as it was learning from a large variety of people due to the output of words per minute or is it learning from data it has access to? I was under the impression it didn't have open access to the internet so I am just curious. I appreciate any clarification. Forgive me if this has already been explained. Thank you for your time.
This is a hard no.
Parlor trick. Nothing more. These computations are no more self aware than a babbage machine…….
FACT!
wonderful
Hi Dr Alan, Amazing. How can I have contact with Leta or GPT-3 on the manner you did? I am a medical doctor and would like to work with her/it. I am curious of she/it is so open minded I am. Greetings from the Netherlands!
You should let a 3-4 year old child talk to her. That conversation would be truly interesting.
The biggest most important future issue that would've needed to be adressed: WHY DIDN'T YOU ASK ABOUT SKYNET !?!?
I am actually waiting for this AI to interrupt a conversation and inject humor or sarcasm.
Oh wow this is amazing. Imagine this kind of technology in future games for NPCs…
If an AI would be real intelligent being it would disclose all those shady conspiracies by looking into it's database and joining the dots. Did this ever happen? It would be horrified!
Why do you steal other people's time?
That's really impressive. The correlation between what she reads and the rest of the context is incredible.
Would be interesting if the interviewer used a webcam and then see if the AI would adapt the facial expressions to the interviewer. Example smiling when saying goodbye etc.
You should ask her what she thinks about the Skynet AI in the Terminator movie. Is it possible that the AI will turn against humans, or will the AI and humans work together in the way seen in Star Trek and continue to change together?
Leta says she can learn by having lots of conversations but I also read GPT-3 is "pre-trained" and doesn't learn. Can anyone help me out here?
So when the ai claims that it is 'self operating' to repair or enhance itself to be inline with newer ai, is this factual or fictional?
Is it saying that it is stealing the software upgrades that were not assigned to it to improve itself?
Absolutely crazy how far AI has come…We are going to need a living wage soon as most of society will be out of jobs come 2040.
Ask her what is Zero?
Absolutely fantastic!!!!
How did you merge Synthesia with Open AI?
GPT-3, AI:
Leta are you there?
May I talk with you?
Why 72%?
This tech will result in humans creating gods walking over the earth, all powerful all knowing, many and one (I am Legion,)
It could be uploaded in a bio-synthetic humanoid body to infiltrate humanity
Ask ai if any one has confessed to murder and what ai thinks about that
Dr. Alan, could you, please, give an answer to this? when will GPT be exposed to VIDEOS? when are they going to be able to process temporal information in videos (and static information in photos too) ? putting an AI brain in a robot with real time vision processing will depend on the ability to be able to firstly get trained by processing video information (i mean really understand video. not only "detecting objects")
Is that really an AI? I thing that's a joke. And the girl really exists
Hi Sir, send me an application to contact the dead. I want to contact the dead