The rise of AI-enabled zero-knowledge threat actors does not merely accelerate attacks—it dismantles the economic and temporal foundations of coordinated vulnerability disclosure. Where traditional researchers relied on months-long windows to notify vendors and allow patching, these actors exploit AI to surface flaws, generate working exploits, and monetize access within days. Verizon’s 2024 DBIR already showed exploitation surging to 31% of initial access vectors; AI compresses that timeline further by automating reconnaissance, payload crafting, and kill-chain orchestration. Smaller supply-chain firms, with weak patching and limited telemetry, become the new entry points precisely because disclosure processes assume a human-scale adversary who needs deep expertise. This dynamic shifts monetization from bug-bounty programs toward real-time underground markets where AI lowers the barrier to weaponization. The responsible disclosure model, built on shared norms between skilled researchers and vendors, cannot survive when intent plus cheap compute replaces skill. Governments and large vendors must now treat rapid exploitation as the baseline and redesign disclosure around automated detection and preemptive mitigation rather than post-discovery coordination.