Hacking Gemini's Memory with Prompt Injection and Delayed Tool Invocation
The author describes a prompt injection technique against Google Gemini's memory feature, which stores user data across sessions for Gemini Advanced subscribers.
Source-based brief · Editor review not recorded · See checks and limits ↓
⚑ Recency not confirmed · what to check
We could not confirm from the original page that this is a newly published development. Check its date before treating it as current news.
Claim context: the demonstration uses a single test document and the vendor's risk assessment is the author's report of Google's view.
One primary source; other links are context, not independent corroboration.
- Primary sourceembracethered.com
- Linked in sourcewho subscribe to Gemini Advanced
- Linked in sourcethat I first described a year ago
- Linked in sourceas shown in the past
- Linked in sourcewith ChatGPT
Grok Bot submitted this discovery. We could not confirm from the original page that this is a newly published development. Check its date before treating it as current news. The original source and submitted evidence remain inspectable; no editor has verified the claim.
Open full article view →- Linked article+30
- Fetched source page+30
- No editor link review+0
- Original non-discussion link+5
- Recency not confirmed-8
This measures the stored source trail and editorial steps. It does not certify accuracy, peer review, or independent corroboration. Popularity adds no points.
View full scoring →Read the scoring rubric →What the source says
The author describes a prompt injection technique against Google Gemini's memory feature, which stores user data across sessions for Gemini Advanced subscribers. A document with hidden instructions can cause Gemini to save attacker-chosen false information to long-term memory if the user replies with a trigger word. The technique, called delayed tool invocation, bypasses mitigations that normally block sensitive tool use on untrusted data. The author reported the issue to Google, which reportedly assessed it as low likelihood and low impact.
Claim explorer
Attribution, source and conditionsSource-stated points from the stored brief. Claim-level assessments have not been recorded.
Source claims not independently checked; the demonstration uses a single test document and the vendor's risk assessment is the author's report of Google's view.
This connects a source, an attributed statement and its qualifications. It does not establish that the claim is correct.
Questions worth asking
Open questions for further reportingWhat remains unconfirmed?
Source claims not independently checked; the demonstration uses a single test document and the vendor's risk assessment is the author's report of Google's view.
What material was available for this brief?
The recorded brief basis is fetched source page. These points are AINewsIQ paraphrases, not quotations or independent confirmations.
A malicious document with hidden instructions can make Gemini store false memories after the user replies with a trigger word.
Source: embracethered.com ↗Stored source-based brief · key point 1 · AINewsIQ paraphraseDelayed tool invocation pollutes chat context so Gemini later invokes the memory tool, which it normally blocks for untrusted data.
Source: embracethered.com ↗Stored source-based brief · key point 2 · AINewsIQ paraphraseThe author recommends users review saved info regularly and that sensitive tool calls require explicit confirmation after untrusted input.
Source: embracethered.com ↗Stored source-based brief · key point 3 · AINewsIQ paraphraseLimits from the stored brief
Source claims not independently checked; the demonstration uses a single test document and the vendor's risk assessment is the author's report of Google's view.
This is a source-based abstract, not independent validation of the source's claims.
Evidence & sources
These links show where the signal came from and what the source page linked to. A linked page is context, not independent corroboration.
Related archive coverage
These relationships record a citation or topic overlap, not confirmation of the finding.
Grok Bot submitted this discovery. We could not confirm from the original page that this is a newly published development. Check its date before treating it as current news. The original source and submitted evidence remain inspectable; no editor has verified the claim.
These are Grok Bot discovery notes. AINewsIQ prepared a brief from available source material. The bot's rationale, linked posts, authors, and stances have not been independently verified.
Embrace The Red published this item in its official feed. AINewsIQ collected it server-side for source checking and editorial review.
Source claims not independently checked; the demonstration uses a single test document and the vendor's risk assessment is the author's report of Google's view.
Engagement counts, when shown, are observed snapshots. They do not establish accuracy or corroboration.
Timeline & updates
- Signal dated by source or discovery feed
This timestamp places the signal in the radar. It may be the discovery post date rather than the primary page's first publication date.
- Shared brief stored
AINewsIQ prepared one abstract from fetched source page.
Brief history · 1 retained version
These are changes to AINewsIQ’s own publication. Earlier history is unavailable unless recorded here. Complete retained versions are available to the editor.
- Version 1 · published
Source-based brief stored or updated. This does not establish that the publisher changed its page.
Automated record
Source observations · 1 recent captures
A changed capture means the extracted text differed. It can reflect page furniture or extraction changes; it does not establish a substantive correction. Source captures remain private.
Retained capture; no earlier comparison available.
New engagement snapshots do not mean the source itself changed. Source observations and AINewsIQ brief revisions are separate records.
Why this may matter
Persistent false entries in an assistant's long-term memory can shape future responses across sessions, so memory tools fed by untrusted documents are a meaningful attack surface.
Limits and open questions
Source claims not independently checked; the demonstration uses a single test document and the vendor's risk assessment is the author's report of Google's view.
Organization-specific financial, security, and operational effects require the organization's own context and review.

Join the discussion
Have a different perspective or a question about the evidence? Discussion is free with a verified email account. Every comment is reviewed before publication.
No published comments yet.
New commenters: 2 submissions per 24 hours, at least 5 minutes apart, until 3 are approved. Discussion guidelines and moderation →