Overview
- Base awards for critical "rogue actions" in flagship products such as Search, Gemini Apps, and core Workspace apps top out at $20,000, with quality and novelty bonuses raising single reports to $30,000.
- Google published tiered categories and payouts, including S1 Rogue Actions ($20,000 flagship), S2 Sensitive Data Exfiltration ($15,000 flagship), and A1 Phishing Enablement ($5,000 flagship).
- In‑scope targets span Google Search on google.com, Gemini Apps on web and mobile, and Workspace core, with AI Studio, Jules, and non‑core Workspace features in a standard tier and other AI integrations at lower tiers.
- Content issues such as hallucinations, hate speech, and copyright concerns are excluded, as are direct prompt jailbreaks and certain cloud surfaces, which should be reported via product feedback or Google Cloud’s VRP.
- Alongside the program, Google introduced CodeMender, an AI agent credited with 72 human‑vetted open‑source security fixes, and noted researchers have earned over $430,000 from AI‑related reports since 2023.