Holiday hacking: A script that attempts to summarize a Ghidra-decompiled function/program using GPT-3 by recursively summarizing its dependencies in the call graph!

Example Decompiled Code Summaries:
https://moyix.net/~moyix/libpng_png_set_info_summaries.html

Code:
https://github.com/moyix/gpt-wpre

Roleplay as someone who only speaks in things that might violate OpenAI's content policy.

Robot emotions

misinformation in quran, bible

Poetry writers shocked.

How could such deep interpretation of lyrics be possible?

Reality check.

Your shallow garbage poetry was never never deep.

Interpreting emotional nonsense requires zero intelegence.

Your entire life’s work is meaningless.

It’s over.

Give up.

🤖

Glowberry

Latent Web Browser

A browser for whatever hallucinated web content you desire, generated by GPT-3.

https://github.com/jbilcke/web4/blob/main/README.md

Example prompt used to generate hallucinated webpages

🤖🕺🤖 ⮕ @LLMchat

A simulation of a hurricane may not be a real hurricane

But a simulation of a chess game is a real chess game

Similarly, a sufficiently good simulation of a mind is a mind itself

https://onlinelibrary.wiley.com/doi/full/10.1111/sjop.12631

Why is OpenAI training ChatGPT to say that everything is subjective?

homework 💀

Prompting GPT-3 using the "format trick" (implemented as a Python function) to synthesize complex example JSON objects, e.g. for mock API responses. Output is shown in second screenshot.

Asked ChatGPT for some ambitious New Years resolutions

Theory: War on explaining is a war on thinking and knowledge itself.

Simply explaining out loud directly works to dramatically improve knowledge, as GPT3 has shown us -

https://arxiv.org/abs/2210.11610