This Prompt Can Make an AI Chatbot Identify and Extract Personal Details From Your Chats

Quick Summary:

Security researchers created an algorithm that turns a malicious prompt into a set of hidden instructions that could send a user's personal information to an attacker.

A spokesperson for Mistral AI says the company welcomes security researchers helping it to make its products safer for users.“Following this feedback, Mistral AI promptly implemented the proper remediation to fix the situation,” the spokesperson says.

The researchers’ algorithm generates an obfuscated version (also above) that has the same meaning to the LLM, but to humans looks like a series of random characters.“Our current hypothesis is that the LLMs learn hidden relationships between tokens from text and these relationships go beyond natural language,” Fu says of the transformation.“It is almost as if there is a different language that the model understands.

---

More Pictures

---

Read Quick SummarySave Time

or

Article Details

Author / Journalist: Matt Burgess

Category: TechnologyEconomy

Markets: United StatesMaliSingapore

Topics: Security / Security News Security Security / Privacy Security / Cyberattacks and Hacks

Source Website Secure: Yes (HTTPS)

News Sentiment: Neutral

Fact Checked: Legitimate

Article Type: News Report

Published On: 2024-10-17 @ 10:30:00 (21 hours ago)

News Timezone: GMT +8:00

News Source URL: wired.com

Language: English

Article Length: 1681 words

Reading Time: 10 minutes read

Sentences: 58 lines

Sentence Length: 29 words per sentence (average)

Platforms: Desktop Web, Mobile Web, iOS App, Android App

Copyright Owner: © Wired

News ID: 23104761

View Article Analysis

About Wired

Main Topics: TechnologyEconomy

Official Website: wired.com

Update Frequency: 18 posts per day

Year Established: 1993

Headquarters: United States

News Last Updated: 12 hours ago

Coverage Areas: United States

Ownership: Independent Company

Publication Timezone: GMT +8:00

Content Availability: Worldwide

News Language: English

RSS Feed: Available (XML)

API Access: Available (JSON, REST)

Website Security: Secure (HTTPS)

Publisher ID: #111

Publisher Details

Frequently Asked Questions