Google’s Project Zero has announced a significant shift in its vulnerability disclosure policy, moving toward a more aggressive timeline for releasing technical details to the public. Under the new framework, dubbed Full X, the research team will disclose discovered vulnerabilities exactly X days after reporting them to a vendor, regardless of whether a patch has been developed or deployed. This policy marks the end of the previous 90-plus-30-day rule, which allowed for a 90-day remediation window followed by a 30-day soak period for patch distribution.
The transition to the new schedule will be phased over the next few years. The current 90-day deadline will remain in effect until February 2, 2027, at which point it will be shortened to 60 days. By January 3, 2028, the disclosure window will contract further to 30 days. Once these deadlines are reached, Project Zero intends to publish technical findings automatically if the time limit expires, even if the vendor has failed to provide a fix.
Project Zero attributes this policy change primarily to the influence of large language models (LLMs) on the threat landscape. The team noted that LLMs have significantly accelerated the ability of attackers to identify vulnerabilities and develop functional exploits. Furthermore, the increased efficiency of these tools raises the probability that multiple parties, including malicious actors, will independently discover the same flaws while they are still under embargo. Consequently, the team argues that the traditional extended delay periods provide a false sense of security while offering diminishing returns for user protection.
The rise in the volume of vulnerabilities discovered by the team has also necessitated a more streamlined approach. Project Zero reported that it is now identifying several vulnerabilities per week in targets where it previously found only a few per year. Managing complex negotiations with vendors regarding whether a bug is considered fixed or whether a grace period applies has become operationally unsustainable. By removing the need to make binary judgments on the status of a fix, the team aims to simplify its internal processes while still providing information on how users can be protected.
The new policy eliminates most previous exceptions and grace periods, though three specific scenarios remain. Vulnerabilities that are already being actively exploited in the wild will continue to be handled under Google’s broader application security disclosure policy. If details of a vulnerability are leaked or published by a third party before the deadline, Project Zero reserves the right to release its own findings immediately. Additionally, the team may grant exceptions for hardware vulnerabilities, acknowledging that these can require more complex remediation efforts, as seen in past cases like Spectre and Meltdown.
Project Zero emphasized that severity ratings will no longer influence the disclosure timeline. The team maintains that it only reports bugs with substantial security impact, and while it acknowledges that vendors may prioritize different risks, a vendor’s internal severity assessment will not delay the public release of information. The policy assumes that all reports meet established standards for reproducibility, readability, and reachability.
During the development of this policy, Project Zero consulted with the vendors it reports to most frequently. These vendors expressed concerns that a rigid, shortened deadline could inadvertently provide attackers with actionable intelligence before fixes are ready. They also worried that such pressure might force them to prioritize Project Zero’s reports over other, potentially more severe vulnerabilities, or lead to a decline in overall remediation efforts if deadlines become impossible to meet.
Despite these concerns, Project Zero maintains that the staged implementation schedule is a realistic response to a changing environment. The team encourages software maintainers to prepare for the 30-day window by developing strategies to protect users that do not rely solely on traditional patching, such as disabling vulnerable code paths. By shifting the focus away from the binary status of a patch, the team intends to prioritize the rapid dissemination of technical information to help the broader security community defend against emerging threats.





