
By 2030, can we convert at least 10% of an AI's weights to C code, enhancing interpretability?
Basic
6
Ṁ1032031
40%
chance
1D
1W
1M
ALL
This question is managed and resolved by Manifold.
Get
1,000and
3.00
Sort by:
@jj does this require training the model first then perfectly converting over in a single step, or can it be re-trained / fine tuned around the C code after?
@jj how about a requirement that the ai is at least better than gpt3 or something? To avoid dumb cases where the ai is really simplistic
Related questions
Related questions
Will researchers extract a novel program from the weights of an LLM into a Procedural/OO programming language by 2026?
24% chance
Will an AI system be able to fully refactor a 10k+ line codebase before 2026 ?
14% chance
Will AI/chatgpt technology be good enough to autoconvert matlab code to runnable python code by EOY 2025?
84% chance
On Dec 31, 2025, will a widely available AI model be able to write a sophisticated 2000 line program?
25% chance
Will there be entry-level AI coders by 2026?
61% chance
Will AI be able to write, compile, and unit test a single .c file to reproduce GPT-2 training from PyTorch code by 2026?
70% chance
Will interpretability be commonplace in physics papers relying on machine learning by the end of 2025?
9% chance
By 2030, will over 50% of software development projects be primarily created by AI, with minimal human coding?
70% chance
In 2029, will any AI be able to construct "reasonably" bug-free code of >= 10k LOC from a natural language specification? (Gary Marcus benchmark #4)
81% chance
Will AI translators be as good as humans before 2030?
89% chance