SpAIware & More: Advanced Prompt Injection Exploits in LLM Applications

#1680549: SpAIware & More: Advanced Prompt Injection Exploits in LLM Applications

Description:	Join Knostic for an engaging and thought-provoking discussion as we delve into the intricate world of advanced prompt injection exploits that target widely used LLM applications, such as Microsoft Copilot, Google Gemini, Google NotebookLM, Apple Intelligence, GitHub Copilot Chat, Anthropic Claude, and many more. Through captivating real-world demonstrations, we will explore the following pressing threats in vivid detail: Misinformation, Phishing, and Scams: Including advanced techniques such as conditional instructions. Automatic Tool Invocation: Exploiting tool integration to escalate privileges, extract sensitive data, or modify system configurations. Data Exfiltration: Leveraging strategies, such as markdown and hidden payloads, to bypass security controls and leak data. SpAIware and Persistence: Manipulating LLM memory for long-term control and persistence. ASCII Smuggling: How LLMs can hide secrets and craft hidden text invisible to users. For each threat category, we will discuss mitigations and show how vendors are addressing these vulnerabilities.
More info:	https://apps.blackhat.com/e/es?s=95530031&e=71880&elqTrackId=CCDC3E7D5DA7C6B39F06ADDC895220E4&elq=ea8a60043b524a42a11288938e477680&elqaid=4313&elqat=1&elqak=8AF573BDE9C3B5793D9EA9B524E85C506B39C4136606996B3F39F877F0878B2701BD

Description:

Join Knostic for an engaging and thought-provoking discussion as we delve into the intricate world of advanced prompt injection exploits that target widely used LLM applications, such as Microsoft Copilot, Google Gemini, Google NotebookLM, Apple Intelligence, GitHub Copilot Chat, Anthropic Claude, and many more. Through captivating real-world demonstrations, we will explore the following pressing threats in vivid detail:
Misinformation, Phishing, and Scams: Including advanced techniques such as conditional instructions.
Automatic Tool Invocation: Exploiting tool integration to escalate privileges, extract sensitive data, or modify system configurations.
Data Exfiltration: Leveraging strategies, such as markdown and hidden payloads, to bypass security controls and leak data.
SpAIware and Persistence: Manipulating LLM memory for long-term control and persistence.
ASCII Smuggling: How LLMs can hide secrets and craft hidden text invisible to users.
For each threat category, we will discuss mitigations and show how vendors are addressing these vulnerabilities.

More info:

https://apps.blackhat.com/e/es?s=95530031&e=71880&elqTrackId=CCDC3E7D5DA7C6B39F06ADDC895220E4&elq=ea8a60043b524a42a11288938e477680&elqaid=4313&elqat=1&elqak=8AF573BDE9C3B5793D9EA9B524E85C506B39C4136606996B3F39F877F0878B2701BD

Date added	May 9, 2025, 10:01 p.m.
Source	Blackhat
Subjects	AI/ML - Artificial Intelligence / Machine Learning / Artificial General Intelligence - AGI - Various Fake News / Hoax News / Deep Fakes / Misinformation / Disinformation - New Reports in Large Language Models (LLMs) Phishing Alerts - Non-Banking PodCasts / Webcast / Webinar / eSummit / Virtual Event etc. Scam/Fraud/Corruption/Swindle/Hoax/Grafts / Kleptocracy News
Venue	May 29, 2025, midnight - May 29, 2025, midnight