"It's well known that artificial intelligence models such as GPT-5.2 improve their performance on benchmark scores as more compute is added. It's a phenomenon known as "scaling laws," the AI rule of thumb that says accuracy improves in proportion to computing power. But, how much effect does computing power have relative to other things that OpenAI, Google, and others bring -- such as better algorithms or different data? To find the answer, researchers Matthias Mertens and colleagues of the Massachusetts Institute of Technology examined data for 809 large language model AI programs. They estimated how much of each benchmark's performance was attributable to the amount of computing power used to train the models."
"They then compared that figure to the amount likely attributable to a company's unique engineering or algorithmic innovation, what they call the "secret sauce," which is sometimes -- but not always -- disclosed. And they compared general improvements in AI across the entire developer community and shared tips and tricks that consistently improve model performance. Their results are reported in the paper "Is there a 'Secret Sauce' in large language model development?", which was posted on the arXiv preprint server."
An analysis estimated the contribution of computing power to benchmark performance across 809 large language models. The analysis compared compute-driven effects to company-specific engineering or algorithmic innovations (termed "secret sauce") and to broadly shared community algorithmic improvements and tricks. Results show total compute had a larger effect on accuracy than proprietary techniques and dominated performance gains, while shared algorithmic improvements also boosted models but less than scale. The conclusion indicates computing power will continue to be a primary driver of AI capability improvements relative to unique proprietary engineering.
Read at ZDNET
Unable to calculate read time
Collection
[
|
...
]