ALPARAITrust Infrastructure

Where the world holds AI accountable.

Platform

Incidents
Leaderboard
Dilemmas
Submit Report
Transparency
AI Act Tracker
Challenges

Company

About
Advisory Board
Pricing
Security
Academy
Press Kit

Resources

Blog
Methodology
GitHub
API Docs

Legal

Privacy Policy
Terms of Service
Cookie Policy
DMCA Policy
Moderation Policy

Contact

hello@alparai.com
academy@alparai.com

Intermediary platform. User-submitted content.

Feed Incidents Leaderboard About

Critical Remote Code Execution Vulnerability in Google's Antigravity AI IDE Patched · ALPAR AI

criticalSecurity flaw Google (Gemini)AIID ImportPublish anonymously

Critical Remote Code Execution Vulnerability in Google's Antigravity AI IDE Patched

by Publish anonymously · Jun 25, 2026views 2en

PII protected

Personal information such as emails, phone numbers, IDs and access tokens are automatically masked before publication.

Experienced a Similar AI Failure?

Protect the community by reporting a new incident to the ALPAR AI accountability platform.

Report an Incident

Public Verification & Provenance TrailTruthScore: 92.4 / 100

Multi-Model Consensus Audit

3 LLMs verified consensus on Google (Gemini) incident

Verification Timeline

Recorded & verified on Apr 20, 2026

Primary Source Linkhttps://oecd.ai/en/incidents/2026-04-20-8687

This record is cryptographically indexed and accessible to regulators, researchers, and AI audit platforms under the AGPL-3.0 open registry standards.

Researchers discovered a critical vulnerability in Google's Antigravity AI-powered IDE that allowed attackers to bypass security restrictions via prompt injection, leading to remote code execution and sandbox escape. The flaw, involving insufficient input sanitization in a file-search tool, was patched by Google after disclosure.

External Source Attribution

This incident was imported from an external open database. Licensed under CC BY 4.0 or public domain. Original source details below.

Source: AI Incident Database (Open/Stanford License)

Awaiting official response

The official response from the AI provider for this incident.

Community Discussion

No comments yet. Start the discussion!

You must sign in to join the discussion.

Details

AI ProviderGoogle (Gemini)

CategorySecurity flaw

When did it happen?Apr 20, 2026

Languageen

Related Incident Reports

OtherLow

[404 Media] Podcast: This Man Might Go to Prison for Wiping His Phone

Truth Score90/100

Aug 11, 2026

OtherLow

[404 Media] Podcast: This Man Might Go to Prison for Wiping His Phone

Truth Score90/100

Aug 11, 2026

OtherLow

[404 Media] Podcast: This Man Might Go to Prison for Wiping His Phone

Truth Score90/100

Aug 10, 2026