braindead_in

I am trying to build react apps with TDD. I've successfully built a todo app with just prompting. Add more features to it now.

> No, because it doesn’t have any underlying model of the world.

Ilya's counter to this reasoning is for next word prediction to work, the model has to 'understand' our world. Otherwise the predictions will be way off. Therefore the human world has been modelled to a degree by GPT.

I haven't yet heard a good counter to that.

Just one mention of India Stack and no mention of iSpirit/Sharad Sharma at all. They were the brains behind it all. I hope they do something with LLMs as well.

Humans, with a lot of help from LLMs.

> natural language programming via LLMs

I wonder how far can this concept itself go. One of the hardest part of software engineering is figuring out what to build and translating that to code. Humans are not particularly good at describing what they want, nor writing code. If human involved is reduced in both, how will software engineering change?

> but a human still has to read that code and make sure that it does what you think it does.

Why not just write tests instead? If you have 100% test coverage, would you care about the code?

I am trying to learn how to build React apps with ChatGPT following the TDD method. Building a todo app is just one single prompt. ChatGPT is pretty good at coming up with test cases given a set of requirements. For the backend, I'm thinking of Supabase.

> To those who believe that new jobs will emerge at meaningful rates to absorb the losses and shocks, what exactly are those new jobs?

Prompt Engineering. The highest paid will be the ones who are able to extract the most value out of LLMs with the least effort.

> I am not a Bhakt,

I am a Bhakt. I became one after I voted for the first time in 2019. I had to choose between Congress and BJP at that time and I decided to go with Congress because I hated Modi for the 2002 riots.

Afterwards, I had a nagging feeling that I had been conned. I am a 90s kid and am staunchly anti-Congress. I have seen first hand how they have ripped off millions in scams and destroyed India. It's almost as if the British loot of India never stopped. But how did I end up voting for corruption then? The con finally dawned on me when I realised that they point at Modi and vilify him not because he's anti-Muslim (he's not) but because he's cleaning up the mess. He is delivering on growth and and transforming India at a rapid pace. He's going to be the Deng Xioaping of India.

However, I was branded as a Hindutva fascist as soon as I voiced my support for him. And I have embraced that term fully. I've studied Hindutva and have turned into a complete Sanatani. My goal in life is now Moksha!

One of my greatest regret in life will be that I didn't vote for him in 2019. I am writing this from the queue of Karnataka assembly elections. I voted for him today.

Combine TDD and Self debugging into a workflow and you almost have a new paradigm of software development where entire applications can be developed with a series of prompts. Software programmers have finally programmed themselves out of jobs! It's kind of poetic justice that LLMs trained on open source code is replacing us.

We should have never listened to Richard Stallman. /s

> Consciousness is awareness of itself. Inner monologue is self referential.

ChatGPT is definitely aware that it exists. In humans, the self-referential inner monologue activates when the mind is idle. The Default Mode Network is responsible for that processing. Can it be that ChatGPT is "thinking" when it's idle as well?

Have we discovered all the emergent abilities of GPT-4?

Is it encoded in genes? Or symbols?

Wow, that blog led me down a rabbit hole. Wonder why Yarvin didn't comment on the societal and political impact of LLMs. Sam Altman semed to be supportive of Democratic Socialism and central planning on Lex Fridman's podcast.

I wonder if some day someone will write a similar story about early days of LLMs.

How can I select a block of code in vim and pass it to this? Maybe I should build a vim plugin for this with ChatGPT.

Live in peace and harmony is not a good principle for AI?

As a fellow believer in the equivalence of Bramhan and Atman, I can't help but think that the only ones amused by the current situation with AI are the Nondualists. Sama seems to be one of them [1]. So, I say to sama, be fearless and invent the future. After all, it's not something new, and in the end, everyone is destined for liberation. Let's enjoy the ride and see where it takes us!

[1] https://twitter.com/sama/status/1607482407770554373

Not to forget the time spent agonizing over naming variables.

Signed up for early access.

Wow, training on a dataset of 12 million hours is quite impressive! I can only imagine the engineering feats required to accomplish that. To put it into perspective, Whisper was trained on 650K hours, Speechmatics' Ursa was trained on 1M hours, and AssemblyAI trained Conformer-1 on 650K hours. I hope Meta is also working on something similar!

That being said, Speaker Diarisation is still a problem that hasn't been fully solved. As of yet, AI hasn't been able to outperform humans in this area.

Interesting comment about consciousness being the fundamental substrate of the world and we are in a dream or simulation. That is exactly the Nondualist position. Did he say Bramhan?

https://youtu.be/L_Guz73e6fw?start=1:08:22

I would like to propose a thought experiment concerning the realm of knowledge acquisition. Given that the scope of human imagination is inherently limited, it is inevitable that certain information will remain beyond our grasp; these are the so-called "known unknowns." In the event that an individual generates a piece of knowledge from this inaccessible domain, how might it manifest in our perception? It is likely that such knowledge would appear incomprehensible to us. Consequently, it is worth considering the possibility that the GPT model is not, in fact, experiencing hallucinations; rather, our human understanding is simply insufficient to fully grasp its output.

I am recording all my calls and building a Q&A bot over the transcripts.

Was this written by ChatGPT? Maybe it found the bug as well, who knows.

The Programmer is dead, long live the Programmer.

To everyone affected by the crisis: Keep calm and build ChatGPT apps.

Where does that leave NLP of everybody can just use prompting + few shot learning using context to build an AI model?

This topic came up in the Sam Harris and Sw Sarvapriyananda podcast recently. Sam mentioned: while the Hindus and Buddhists kept debating, the Muslims won.

Can it be used for Whisper?