New RSA attack could sharply lower cracking costs, researchers say
Researchers from UC San Diego and Inria Nancy have proposed an attack that significantly lowers the computation needed to factor RSA keys. If validated, the method could reduce operation counts for 1024- to 4096-bit keys to levels potentially within reach of well-resourced actors. RSA remains widely deployed despite being deprecated in favor of newer algorithms.
Researchers at UC San Diego and Inria Nancy have outlined a factoring method that, if peer review confirms it, would cut the work needed to break RSA keys. Their figures drop from 2^80, 2^112, 2^128, and 2^144 operations to 2^65, 2^90, 2^105, and 2^119 for 1024-, 2048-, 3072-, and 4096-bit keys.
The technique applies only to textbook RSA, not the common padded form used for most web certificates. It still matters for Cloudflare Privacy Pass, iCloud Private Relay, Private Cloud Compute, and PKCS #11 devices such as smart cards, tokens, HSMs, and TPMs. Attackers would need enormous numbers of oracle queries, and the algorithms remain unoptimized.
If validated, this research could shift risk for organizations and users relying on textbook RSA, including privacy relays, hardware tokens, HSMs, and TPMs. Well-resourced actors may eventually gain more ability to inspect protected traffic or compromise signing workflows. Most everyday website and remote-access encryption would likely remain unaffected because padded RSA and newer algorithms are common. The attack's heavy query requirements and lack of optimization mean real-world impact may stay limited for now.