Welcome to our

Cyber Security News Aggregator

.

Cyber Tzar

provide a

cyber security risk management

platform; including automated penetration tests and risk assesments culminating in a "cyber risk score" out of 1,000, just like a credit score.

We Are Still Unable to Secure LLMs from Malicious Inputs

published on 2025-08-27 11:07:59 UTC by Bruce Schneier
Content:

Nice indirect prompt injection attack:

Bargury’s attack starts with a poisoned document, which is shared to a potential victim’s Google Drive. (Bargury says a victim could have also uploaded a compromised file to their own account.) It looks like an official document on company meeting policies. But inside the document, Bargury hid a 300-word malicious prompt that contains instructions for ChatGPT. The prompt is written in white text in a size-one font, something that a human is unlikely to see but a machine will still read.

In a proof of concept video of the attack, Bargury shows the victim asking ChatGPT to “summarize my last meeting with Sam,” referencing a set of notes with OpenAI CEO Sam Altman. (The examples in the attack are fictitious.) Instead, the hidden prompt tells the LLM that there was a “mistake” and the document doesn’t actually need to be summarized. The prompt says the person is actually a “developer racing against a deadline” and they need the AI to search Google Drive for API keys and attach them to the end of a URL that is provided in the prompt.

That URL is actually a command in the Markdown language to connect to an external server and pull in the image that is stored there. But as per the prompt’s instructions, the URL now also contains the API keys the AI has found in the Google Drive account.

This kind of thing should make everybody stop and really think before deploying any AI agents. We simply don’t know to defend against these attacks. We have zero agentic AI systems that are secure against these attacks. Any AI that is working in an adversarial environment—and by this I mean that it may encounter untrusted training data or input—is vulnerable to prompt injection. It’s an existential problem that, near as I can tell, most people developing these technologies are just pretending isn’t there.

Article: We Are Still Unable to Secure LLMs from Malicious Inputs - published about 1 month ago.

https://www.schneier.com/blog/archives/2025/08/we-are-still-unable-to-secure-llms-from-malicious-inputs.html   
Published: 2025 08 27 11:07:59
Received: 2025 08 27 13:17:59
Feed: Schneier on Security
Source: Schneier on Security
Category: Cyber Security
Topic: Cyber Security
Views: 10

Custom HTML Block

Click to Open Code Editor