13-04-2023, 12:19 AM
Basically if you don't know who Sam Altman is, he's the Main Founder of OpenAI
![[Image: 230213-sam-altman-se-1051a-49a88b.jpg]](https://media-cldnry.s-nbcnews.com/image/upload/rockcms/2023-02/230213-sam-altman-se-1051a-49a88b.jpg)
this guy didn't invent anything new.
nothing.
So what exactly did he do?
Now he came across a research paper from 2017. the research paper was done by Google Scientists.
What was the paper about?
Transformer.
An architecture for Natural Language Processing Neural Nets
(link: https://ai.googleblog.com/2017/08/transformer-novel-neural-network.html)
Basically this was Google saying "hey we invented this thing called Generative Pre-Trained Transformers and we dont know how powerful it is but we did it I guess"
After this research paper released, OpenAI was in turmoil as Elon Musk had left his seat at the board of OpenAI, citing "conflict of interest", but basically that was just the rest of the board disagreeing to let Musk take over of the company's leadership and Musk leaving in rage cuz he felt Google mogged OpenAI to dirt.
Anyways, after that, Sam Altman's fellow employees looked into transformers paper by google. they were mindblown by how much google underrated it and decided to work on its flaws.
they did and they released this paper: https://cdn.openai.com/research-covers/language-unsupervised/language_understanding_paper.pdf
And yeah that lead to OpenAI releasing the GPT engine and subsequently GPT2, and of course last year, they dropped GPT-3 which was an insane moment.
So this is why Engineers mog Scientists.
Scientists can create cool theories and models. but what good is creating these theories and models if its not put to any use?
Google Scientists created Transformers but they didn't really put it to terrific use like OpenAI did.
Sam Altman saw this vision and he greenlit it.
this guy didn't invent anything new.
nothing.
So what exactly did he do?
Now he came across a research paper from 2017. the research paper was done by Google Scientists.
What was the paper about?
Transformer.
An architecture for Natural Language Processing Neural Nets
(link: https://ai.googleblog.com/2017/08/transformer-novel-neural-network.html)
Basically this was Google saying "hey we invented this thing called Generative Pre-Trained Transformers and we dont know how powerful it is but we did it I guess"
After this research paper released, OpenAI was in turmoil as Elon Musk had left his seat at the board of OpenAI, citing "conflict of interest", but basically that was just the rest of the board disagreeing to let Musk take over of the company's leadership and Musk leaving in rage cuz he felt Google mogged OpenAI to dirt.
Anyways, after that, Sam Altman's fellow employees looked into transformers paper by google. they were mindblown by how much google underrated it and decided to work on its flaws.
they did and they released this paper: https://cdn.openai.com/research-covers/language-unsupervised/language_understanding_paper.pdf
And yeah that lead to OpenAI releasing the GPT engine and subsequently GPT2, and of course last year, they dropped GPT-3 which was an insane moment.
So this is why Engineers mog Scientists.
Scientists can create cool theories and models. but what good is creating these theories and models if its not put to any use?
Google Scientists created Transformers but they didn't really put it to terrific use like OpenAI did.
Sam Altman saw this vision and he greenlit it.
