Artificial intelligence systems accept input in the form of images, audios, texts, etc. As a result, filtering, handling, and detecting malicious input and behavior has become more complicated. Cyber security is a top priority for companies around the world. The increase in the number of AI security papers from just 617 in 2018 to over 1500 in 2020 (an increase of nearly 143% according to an Adversa report) is evidence of the growing importance of cybersecurity.
Microsoft recently announced the release of Counterfit – a tool for testing the security of AI systems – as an open source project. The tech giant’s latest move could potentially be a giant leap in building a resilient AI ecosystem.
Tech behind it
The tool is particularly suitable for penetration tests and the red teaming of the AI systems. With pre-installed published attack algorithms, the tool is already trained on the worst attacks. In addition, security experts from existing …