In response to a rapidly evolving cybersecurity landscape fundamentally transformed by the advent of artificial intelligence and automated exploitation techniques, Google Project Zero is dramatically overhauling its vulnerability disclosure policy. The elite elite security research team is significantly shrinking its traditional bug disclosure windows, aiming to force the broader technology industry to accelerate its patching processes and modernize its remediation infrastructure.
For years, the standard operating procedure for Google Project Zero was widely understood across the software development and cybersecurity industries. Under that long-standing policy, technology vendors were granted a standard window of 90 days to research, develop, and deploy patches for reported vulnerabilities before any details were made public. Additionally, vendors could request and receive a 30-day grace period for rolling out these patches if they could demonstrate active progress, effectively stretching the total timeline to 120 days in certain circumstances.
However, that era of extended grace periods is drawing to a close. Natalie Silvanovich, a security engineer at Google Project Zero, announced that starting November 13, the traditional 30-day grace period is officially eliminated from the books. Under the newly introduced policy, which Project Zero has designated as the "Full X" policy, the team will publish comprehensive vulnerability details precisely 90 days after initially reporting the security flaws to the respective vendors. Crucially, this disclosure will occur regardless of whether a vendor has successfully released a fix, and regardless of where they are in their patch rollout cycle.
This shift marks only the beginning of a much more aggressive timeline designed to outpace the speed at which malicious actors operate, particularly as automated tools and AI-driven vulnerability discovery reshape the threat ecosystem. Project Zero has laid out a phased roadmap for further tightening these constraints over the coming years.
Beginning February 2, 2027, Project Zero will further compress its timeline, publicly disclosing vulnerabilities just 60 days after initial notification. The pressure will intensify even further on January 3, 2028, when the disclosure window will shrink to a mere 30 days. By accelerating these timelines, the security research team hopes to fundamentally alter how software vendors prioritize security debt and allocate engineering resources toward vulnerability remediation.
Alongside these tightened deadlines, Project Zero is systematically removing most of the "special cases" and exceptions that previously allowed vendors extra time under specific circumstances. Moving forward, very few exceptions will be tolerated. The primary exceptions that will remain intact are security defects that are already being actively exploited in the wild—which maintain a rapid 7-day disclosure window in accordance with Google’s broader vulnerability disclosure policy—and vulnerabilities that have already been made public by third parties or independent researchers.
Despite the rigidity of the new framework, some flexibility remains built into the process. Silvanovich noted that if a vendor explicitly requests it, Project Zero may choose to publish details regarding a security hole ahead of the standard deadline. Furthermore, extremely serious vulnerabilities of monumental proportions may still be granted rare exceptions. Project Zero has exercised this discretion on rare, historically significant occasions in the past, such as the disclosure of the monumental Spectre and Meltdown hardware flaws. Silvanovich confirmed that this cautious approach to once-in-a-generation systemic threats will not change under the new policy.
Explaining the rationale behind this phased and demanding transition, Silvanovich pointed to several critical industry concerns that the new policy attempts to address. Chief among them is the persistent risk that end users remain exposed to unpatched vulnerabilities for extended periods while vendors deliberate. Furthermore, Project Zero observed perverse incentives in the old model, where vendors might rush to fix newly reported flaws under the pressure of an impending public deadline while entirely overlooking older, previously reported bugs that actually posed significantly higher severity risks. There was also the ongoing challenge that certain vendors simply could not or would not meet the demands of the old policy without external pressure.
"We decided on a policy that encourages vendors to improve remediation time on a schedule that is challenging but realistic, with the hope that this will lead to vendors broadly improving their remediation processes and infrastructure in a way that eventually enables them to fix all reported vulnerabilities quickly," Silvanovich noted.
Addressing the mechanics of their research and the types of flaws they handle, Silvanovich emphasized that Project Zero exclusively targets and reports vulnerabilities that possess substantial severity and a clearly demonstrated impact. She also pointed out an important nuance in threat modeling: a vulnerability’s inherent severity strictly affects its potential impact on users and systems, but it does not inherently dictate the likelihood that malicious attackers will discover it independently. In the modern age of automated scanning and AI-assisted exploitation, the speed at which attackers discover flaws has converged with, and often outpaced, the defensive response.
To ensure fairness in an otherwise uncompromising system, Silvanovich stressed that Project Zero holds itself to exceptionally high internal standards regarding the reproducibility, readability, reachability, and reported security impact of every single vulnerability its researchers uncover and report. The entire premise of this aggressive new disclosure policy relies on the assumption that the bug reports handed over to vendors meet this rigorous quality threshold. She acknowledged that it remains entirely reasonable for vendors to treat low-quality bug reports differently, but insisted that Project Zero’s own high standards justify the swift timelines being imposed.
As the tech industry grapples with an unprecedented surge in automated and AI-driven vulnerability discovery, traditional patching models are facing an existential stress test. By shrinking disclosure windows and removing comfortable grace periods, Google Project Zero is forcing a reckoning across the software ecosystem, signaling that the luxury of time is no longer available in modern cybersecurity.

