Intel Makes ControlFlag Open-Source For Helping To Detect Bugs In Code

Last year Intel announced ControlFlag as a machine learning tool for helping to uncover bugs within code. ControlFlag promised impressive results after being trained on more than one billion lines of code and at the end of 2020 was already being used internally on Intel's code-bases from firmware to software applications. We hadn't heard anything more about ControlFlag this year... Until today. Intel has now made ControlFlag open-source for helping to autonomously detect more programming bugs.

ControlFlag is described by Intel Labs engineers as "A Self-supervised Idiosyncratic Pattern Detection System for Software Control Structures" but what it comes down to is basically from mining patterns within C/C++ code-bases of many open-source projects, it detects anomalous patterns in user's code.

ControlFlag can be built for Linux and macOS systems. Intel has made training data available that they generated from 6,000 open-source GitHub repositories. From there it's possible to easily scan C/C++ code in looking for potential anomalies.

Intel reported today that not only is the open-source code now public but it has already "found hundreds of confirmed software defects in proprietary, production-quality software." More details can be found via this LinkedIn post by Intel's Justin Gottschlich.

The open-source ControlFlag code is hosted under Intel Labs on GitHub.