- Security experts uncovered a vulnerability in Apple Silicon chips, exploited by hackers to pilfer user data via the Data Memory-Dependent Prefetcher (DMP).
- Named “GoFetch,” the attack allows hackers to manipulate memory access patterns to extract sensitive data, posing challenges for Apple’s short-term fixes due to its core chip placement.
- Any mitigation measures Apple implements may impair performance, given the central chip location of the vulnerability.
Security experts discovered a security vulnerability in Apple Silicon chips, exploited by hackers to steal user data. The flaw, found in the Data Memory-Dependent Prefetcher (DMP), enables hackers to steal encryption keys and access user data.
Also read: Apple co-founder Steve Wozniak wins fake video lawsuit v YouTube
Also read: Apple in talks to let Google’s Gemini power iPhone AI features
It will affect the prefetching process and access sensitive user data
The DMP, also known as an indirect memory prefetcher, predicts the memory addresses of data likely to be accessed by running code. Hackers can manipulate existing access patterns to predict the next data location, affecting the prefetching process and accessing sensitive user data. Researchers named this attack “GoFetch.”
They confirmed that hackers can disguise data as a pointer, tricking the DMP into treating it as an address and pulling it into the cache. While this attack doesn’t immediately crack encryption keys, repeated attempts can eventually retrieve the keys.
The GoFetch attack operates with the same user permissions as many third-party macOS applications, requiring no root access, which lowers the barrier to launching the attack. In tests, researchers’ applications extracted 2048-bit RSA keys in under an hour and 2048-bit Diffie-Hellman keys in just over two hours.
It is difficult to fully repair in the short term
Apple faces the challenge that this vulnerability exists in the core of its Apple Silicon chips, making it difficult to fully repair in the short term. Additionally, any mitigation measures deployed by Apple would increase the workload required for executing operations, thereby affecting performance.
