AI models project varying risks to Bitcoin cryptography from quantum computing

Four artificial intelligence models have provided varying estimates regarding the risk of Bitcoin’s cryptographic security being compromised within the next 20 years. The analysis suggests that while short-term risks through 2028 remain extremely low (0% to 2%), the long-term threat increases significantly as quantum computing technology advances.

Estimates for the 20-year risk horizon vary widely among the models: ChatGPT-6 Astra estimated a 76% chance of a cryptographic break, Claude Opus 5.5 projected 66%, Grok estimated 60%, and Deepseek Deepthink provided the most conservative estimate at 34%. The primary driver for this projected vulnerability is the potential for quantum computers to use Shor’s algorithm to derive private keys from exposed public keys.

Industry experts and organizations are already addressing these concerns. Google Quantum AI has noted that the resources required to crack 256-bit elliptic curve cryptography may be lower than previously anticipated. Meanwhile, the National Institute of Standards and Technology (NIST) has released post-quantum cryptography standards, with experts urging organizations to begin transitioning to these new protocols immediately, as replacing vulnerable cryptographic systems can take up to two decades.