OpenAI
2 HN recommenders · recommended in 5 months, 2021–2025
29:43 · 913K views · Aug 10, 2021
OpenAI's 2021 live demo of Codex translating natural language into code, from Hello World to building a game and a Microsoft Word plugin.
HN points to it as an early, impressive demonstration that LLMs could write working code.
6 comments from Hacker News
I loved Openai's codex JavaScript game making live demo(1) where they actually created a fully functioning game in JavaScript using just plain old english. Kind of changed my whole view about programming and it's future. (1) this video
Here's a pretty impressive demo. this video I think they said it handles something like 37% of requests. Btw, don't watch it if your worried at losing your job to a computer.
That's a rather silly strawman. OpenAI demonstrated Codex - a version of GPT-3 which could write Python and JS code - in 2021, only a year after the first GPT-3 release. Here's a demo of Codex doing MS Word tasks using Python code: [LINK] Live coding demo: this video There's also a demo of doing some data processing using Codex. The idea of using LLM to write code is rater obvious, and many people talked about it around GPT-3 or even GPT-2 release. People know that many logic tasks require search and it's rather silly to use LLM to do that. If it can generate any text it can generate Prolog code or a specification for SAT solver, and then dedicated efficient tool can handle the computation. Gary Marcus is so detached from actual research that he might have genuinely missed the 2021 version of Codex. So he might truly believe that they added Python interpreter "quietly", and it's not the big announcement of 2021. This guy is known as a clown in the industry, and for a very good reason... If he was an actual researcher he would have jumped on the opportunity to make an actual neurosymbolic demos when first language models were released - you don't actually need to wait GPT-3 to do that. But he preferred to relish in writing books claiming that the entire industry is wrong instead of actually building something.
Fair point on sensationalism, I think it's a useful hook to get folks reading and talking. There is a "Nuance" section at the bottom where I qualify most of the rah rah language. That being said, I disagree that it is that far from actually working. Check out the codex demo here: this video This is a year old and we've made a lot of progress since then. The model is certainly capable of writing non pseudo-code, although I started hitting rate limits before I got it all the way there in my example. Edit (typos)
Yea the code on the top level is very cartoonish and almost comical. But the model can definitely produce running code. Chat GPT is a similar model that runs codex, see the demo here: this video One point I try to get across is that once you drill deep enough into the call stack of implementing subroutines, you will start just asking it to produce code not explicitly related to the top level mission of overthrowing humanity.