Researchers have developed a technology to swap the insecure foundations of global software with a safer alternative, outperforming artificial intelligence by guaranteeing “mathematical correctness” in the translation process.
The C programming language has underpinned critical systems — including operating systems and web browsers — since the 1970s. However, its age means it suffers from structural limitations that continuously cause severe bugs and security vulnerabilities.
To address this, the tech industry is shifting toward Rust, a modern programming language developed in 2015 that automatically detects and prevents bugs before the software runs. The push is so urgent that the US White House released a report in 2024 recommending the discontinuation of C, while the Defense Advanced Research Projects Agency (DARPA) declared Rust the “core alternative” for resolving security issues.
Automated massive task
While tech companies have raced to use AI large language models (LLMs) to automate the massive task of translating billions of lines of legacy code, these models often make mistakes or “hallucinate” errors.
The team at the Korea Advanced Institute of Science and Technology (KAIST) pioneered a different approach. Instead of relying on AI probability, they developed a core technology that mathematically proves the accuracy of the conversion, ensuring no new errors are introduced into critical infrastructure.
“The conversion technology we developed is an original technology based on programming language theory, and its biggest strength is that we can logically prove the ‘correctness’ of the conversion,” said Dr Jaemin Hong, the paper’s first author. “While most research relies on Large Language Models (LLMs), our technology can mathematically guarantee the correctness of the conversion,” he added.
