Google adds AI verification and ransomware safeguards to Workspace for Education

Google has introduced a set of security and integrity updates for Workspace for Education, adding tools to verify AI-generated media, detect ransomware and give school administrators more granular control over online activity.

The changes were announced at the Bett conference and reflect growing concern among schools about both the misuse of generative AI and the rising risk of cyberattacks on education systems.

One of the most prominent additions is AI media verification. Google said it is rolling out SynthID watermarking for images and video that are created or edited using Google’s AI tools. SynthID embeds a digital marker that is invisible to the eye but can be detected by software, helping to identify content that has been generated or altered by AI.

Verification will be available through the Gemini app, where users can upload images or video and ask a simple question: “Is this AI-generated?” For teachers and students, this is intended to support media literacy by making it easier to distinguish between authentic material and content produced by AI, without requiring technical expertise.

The update also addresses a more immediate operational risk: ransomware. Google said Drive for desktop will now pause file syncing automatically if ransomware is detected on a device. Ransomware is a type of malicious software that encrypts files and demands payment to unlock them, and schools have increasingly been targeted because of their limited IT resources.

When suspicious activity is detected, both administrators and users will receive alerts. Affected users will also be able to restore multiple files to versions created before the infection, helping schools recover data without resorting to backups or paying ransoms.

For IT teams, Google is expanding visibility and control. A new security operations, or SecOps, data connector allows Workspace activity logs to be forwarded directly into the Google SecOps platform. This feature is available to Education Plus and Education Standard customers and is designed to help security teams monitor activity, investigate incidents and respond more quickly when something goes wrong.

In practical terms, this means events such as logins, file access and sharing can be analysed alongside other security data, rather than sitting in separate systems. For schools with stretched IT staff, centralising this information can make it easier to spot unusual behaviour before it becomes a serious incident.

Google has also added more granular administrator controls for Google Meet live streams. Schools can now better manage who can access or interact with streamed sessions, reducing the risk of disruption or unauthorised viewing.

Alongside this, the company introduced an opt-in “Adaptive” meeting type. This setting enables the new access controls for selected meetings while leaving existing setups unchanged, allowing schools to adopt the changes gradually rather than overhaul all meetings at once.

Related reading

• Google Vids crowns first education film festival winners
• Google.org backs Sundance push to put AI literacy in filmmakers’ hands
• LatticeFlow AI buys AI Sonar to expand enterprise oversight of artificial intelligence

The updates come as schools face a dual challenge. On one hand, generative AI tools are becoming part of everyday learning and teaching, raising questions about authenticity, assessment and trust. On the other, education remains one of the sectors most exposed to cyberattacks, with limited budgets to absorb disruption.

Google said it remains committed to strengthening security and AI integrity in Workspace for Education and plans to introduce further updates in the coming months. The message is that as digital tools become more embedded in classrooms, the infrastructure around them must evolve to protect both learning and learners.

The Recap

• Google adds AI verification and security features to Education suite.
• Drive for desktop pauses syncing on ransomware detection to protect files.
• Company plans to expand verification to audio and non-Google models.