Summary

Top Articles:

  • LLM Agents can Autonomously Exploit One-day Vulnerabilities
  • LLM spews nonsense in CVE report for curl
  • Multi-modal prompt injection image attacks against GPT-4V
  • Training AI Using ‘Pirated’ Content Can Be Fair Use, Law Professors Argue
  • GitHub Copilot Chat: From Prompt Injection to Data Exfiltration
  • Prompt Injection on Bing Chat triggered by search content
  • HuggingFace hacked - Space secrets leak disclosure
  • The Burn Notice, Part 1/5 — Revealing Shadow Copilots

GitHub Copilot Chat: From Prompt Injection to Data Exfiltration

Published: 2024-06-17 12:58:34

Popularity: 4

Author: embracethered.com via kivikakk

Keywords:

  • security
  • ai
  • 🤖: ""Code red alert""

    Comments

    ...more

    LLM spews nonsense in CVE report for curl

    Published: 2024-01-02 22:23:13

    Popularity: 35

    Author: skeptrune@users.lobste.rs (skeptrune)

    Keywords:

  • security
  • ai
  • Comments

    ...more

    Multi-modal prompt injection image attacks against GPT-4V

    Published: 2023-10-14 03:44:10

    Popularity: 25

    Author: simonw@users.lobste.rs (simonw)

    Keywords:

  • security
  • ai
  • Comments

    ...more

    Prompt Injection on Bing Chat triggered by search content

    Published: 2023-03-01 03:36:34

    Popularity: None

    Author: carlmjohnson@users.lobste.rs (carlmjohnson)

    Keywords:

  • security
  • ai
  • Comments

    ...more

    LLM Agents can Autonomously Exploit One-day Vulnerabilities

    Published: 2024-04-24 00:02:44

    Popularity: 63

    Author: arxiv.org via thombles

    Keywords:

  • security
  • ai
  • Comments

    ...more

    HuggingFace hacked - Space secrets leak disclosure

    Published: 2024-06-01 11:38:06

    Popularity: None

    Author: huggingface.co via mark

    Keywords:

  • security
  • ai
  • 🤖: "AI losing data"

    Comments

    ...more

    The Burn Notice, Part 1/5 — Revealing Shadow Copilots

    Published: 2025-02-24 12:23:48

    Popularity: None

    Author: Dor Attias

    Keywords:

  • copilot-studio
  • ai-agent
  • cybersecurity
  • ai
  • data-leak
  • 🤖: "Shadowy secrets"

    How We Extracted Financial Data from a Multi-Billion-Dollar Company Continue reading on Medium »

    ...more

    Training AI Using ‘Pirated’ Content Can Be Fair Use, Law Professors Argue

    Published: 2025-04-02 20:10:14

    Popularity: 21

    Author: Ernesto Van der Sar

    Keywords:

  • AI
  • ai
  • fair use
  • Meta
  • 🤖: "Copy cat!"

    A group of prominent intellectual property law professors has weighed in on the high-stakes AI copyright battle between several authors and Meta. In an amicus brief, the scholars argue that using copyrighted content as training data can be considered fair use under U.S. copyright law, if the goal is to create a new and 'transformative' tool. This suggests that fair use could potentially apply to Meta's training process, even if the underlying data was obtained without permission. From: TF, for the latest news on copyright battles, piracy and more.

    ...more

    end