Claude reveals ECDSA curve origins (anthropic.com)

Article summary

The origins of the ECDSA curve seeds have been revealed, showing they were generated using the C library's lrand48 random-number generator, seeded with a Unix timestamp. This discovery resolves a long-standing mystery in the cryptographic community and provides evidence that the curves were chosen without intentional weaknesses. The seeds were found using a model called Claude, which was able to recognize the output of the lrand48 generator and recover the initialization state. The discovery has implications for the use of ECDSA in various cryptographic applications.

Main themes

• ECDSA curve origins
• Cryptography
• Random number generation
• Unix timestamps
• Cryptographic security
• Model-based discovery

What commenters say

• The revelation of the ECDSA curve seeds' origin resolves a long-standing conspiracy theory and increases trust in the cryptographic community.
• The lack of transparency from NIST about the seeds' origin was unnecessary and damaged trust in the independent cryptography community.
• The use of models like Claude to discover the seeds' origin highlights the potential for models to contribute to cryptographic research and security.
• The discovery has significant implications for the use of ECDSA in applications like Bitcoin and Ethereum, allowing for more confident implementation of the curves.
• Some commenters are concerned that the capabilities of models like Claude are not equally accessible to the public and may widen the gap between those with access to advanced models and those without.
• Others argue that restricting access to models and information is not an effective way to prevent malicious use, and that controls on equipment and reagents are more effective in preventing biological weapon development.
• The ability of models to contribute to cryptographic research and security is seen as a positive development, but also raises concerns about the potential for models to be used for malicious purposes.
• The discovery of the seeds' origin is seen as a testament to the power of models and machine learning in resolving long-standing mysteries and contributing to scientific progress.