AI Uncovers Flaws in Deprecated Encryption Standard
TREE NEWS reports: Google DeepMind researchers have used a human-guided AI agent to uncover two significant weaknesses in the obsolete 1024-bit RSA cryptosystem, potentially lowering the computational cost required to break it. The findings, disclosed by Pushmeet Kohli, Vice President of Science at Google DeepMind, include a method for forging digital signatures and a new algorithm that reduces the average cost of factoring multiple keys of the same bit length. The research only affects 1024-bit RSA, which has been phased out; modern protocols using 2048-bit or longer keys remain secure.
Implications for Blockchain and Crypto Security
While the affected RSA variant is no longer used in contemporary cryptographic protocols, the research carries notable implications for the cryptocurrency and blockchain industry. Many legacy systems, including older hardware wallets and certain enterprise blockchain implementations, may still rely on 1024-bit RSA for key management or authentication. The ability to forge signatures could undermine trust in these systems, especially if they are still in operation.
More broadly, the use of AI to discover cryptographic weaknesses signals a new frontier in security research. AI agents can potentially explore complex mathematical spaces more efficiently than traditional methods, accelerating the discovery of vulnerabilities. This raises concerns about the future of cryptographic standards, as adversaries may leverage similar techniques to attack current systems.
Industry Response and Future Outlook
The crypto industry has largely moved to elliptic curve cryptography (ECC) and other modern algorithms, but the research serves as a reminder of the importance of staying ahead of potential threats. Projects that still rely on RSA, particularly in cross-chain bridges or legacy smart contracts, should consider migrating to quantum-resistant algorithms. Additionally, the integration of AI in security auditing could become a double-edged sword: while it helps defenders find flaws, it also empowers attackers.
Looking forward, the incident underscores the need for continuous cryptographic agility. As AI tools become more accessible, the pace of vulnerability discovery may increase, pushing the industry toward more robust and adaptive security frameworks. Regulators and standards bodies may also need to update guidelines to account for AI-driven cryptanalysis, ensuring that both legacy and emerging systems are adequately protected.
