Dragonfly managing partner Haseeb Qureshi said on X that security is increasingly becoming a cost competition between attackers using frontier AI to scan for vulnerabilities and defenders paying for AI-based hardening. He said Claude was able to find the Coldcard-related flaw in eight minutes, while GLM reproduced it in 20 minutes, which in his view suggested Coldcard had made almost no investment in AI security hardening. Qureshi estimated that the discovery cost of reproducing the attack with GLM 5.2 was only about $2, meaning an AI model could identify the vulnerability for roughly that amount. He warned that small companies will struggle to survive in security-sensitive product categories and argued that any security-sensitive product should undergo AI security scanning on every release.
Dragonfly managing partner Haseeb Qureshi said in a post on X that the key issue in security has become a cost competition: the cost for attackers to use frontier AI to scan for vulnerabilities versus the amount defenders spend on AI-based security hardening.
Qureshi said Claude could find the flaw in eight minutes, while GLM could reproduce it in 20 minutes. In his reading, that showed Coldcard had invested almost nothing in AI security hardening.
He estimated that the discovery cost of reproducing the attack with GLM 5.2 was only about $2, meaning the vulnerability could be found with AI for roughly $2.
Qureshi also warned that small companies will have a hard time surviving in security-sensitive product sectors. He said any security-sensitive product should be put through AI security scanning on every release.
This article was originally published by Bit.Fan. For more cryptocurrency news and market insights, visit www.bit.fan. Disclaimer:
The market information, project data, and third-party content displayed on this platform are for industry information sharing only and do not constitute any form of investment advice or return commitment.
Cryptocurrency trading carries high risks. Users should fully assess their risk tolerance and make independent decisions. All profits, losses, and legal responsibilities are borne by the users themselves.
