Google Project Zero, the security research division responsible for identifying critical vulnerabilities in major software and hardware platforms, has announced a significant shift in its vulnerability disclosure policy. The team is moving away from its long-standing 90-day disclosure window, plus a potential 30-day grace period, in favor of a compressed timeline. This decision is driven by the team's assessment that artificial intelligence is now accelerating both the discovery of security flaws and the development of exploits by malicious actors.
The transition to this new policy began on November 13, 2026, with the immediate removal of the 30-day extension previously granted to vendors. Following this initial change, the standard disclosure deadline will be reduced to 60 days starting February 2, 2027. The final phase of this policy adjustment will occur on January 3, 2028, when the window will shrink to a fixed 30-day period. Under these new rules, Project Zero will release technical details regarding a vulnerability once the deadline expires, regardless of whether the affected vendor has successfully produced or deployed a patch.
Project Zero, which has historically uncovered high-impact flaws such as those affecting the .NET Framework, Chrome’s JavaScript engine, and various Qualcomm chipsets, maintains that the current threat landscape has changed fundamentally. The team argues that AI tools have lowered the barrier for attackers to reverse-engineer security patches as soon as they are released, effectively shortening the window of safety that traditional disclosure policies were designed to provide. By forcing a faster turnaround, the team aims to pressure vendors to improve their internal remediation workflows and accelerate the deployment of security updates to end users.
The decision to implement these changes gradually was made in response to feedback from various software and hardware vendors. During consultations, companies expressed concern that an immediate shift to a 30-day deadline could inadvertently harm users. Vendors warned that such a rapid transition might force them to prioritize Project Zero reports over other, potentially more critical vulnerabilities discovered by different researchers, thereby creating an unbalanced security posture.
Google acknowledged these concerns, noting that while the utility of long embargo periods is diminishing due to the increased likelihood of independent discovery by attackers, there remains a short-term value in allowing vendors time to adapt their systems. The gradual rollout is intended to provide organizations the necessary runway to refine their vulnerability management processes before the 30-day requirement becomes the standard.
Despite the move toward shorter, fixed deadlines, Project Zero has confirmed that it will maintain certain exceptions to the policy. The team will continue to exercise discretion in cases where a vulnerability is already being actively exploited in the wild, if the details of the flaw or its exploitation have already been made public by other parties, or if a vendor specifically requests an earlier disclosure.
Furthermore, Google has reserved the right to withhold information beyond the standard deadlines if it deems such action necessary for security reasons. The team cited the historic handling of the Spectre and Meltdown vulnerabilities—where disclosure was delayed for more than 120 days—as a precedent for when the standard timeline may be bypassed.
This policy shift marks a departure from the industry-standard 90-day disclosure model that Project Zero helped popularize. By mandating faster disclosure, the team is signaling that the traditional approach to vulnerability management is increasingly incompatible with the speed at which modern, AI-assisted threats operate. The success of this new policy will depend on the ability of vendors to meet these tighter deadlines without compromising the quality or stability of their security patches.





