Tag: concept
-
Slashdot: One Long Sentence is All It Takes To Make LLMs Misbehave
Source URL: https://slashdot.org/story/25/08/27/1756253/one-long-sentence-is-all-it-takes-to-make-llms-misbehave?utm_source=rss1.0mainlinkanon&utm_medium=feed Source: Slashdot Title: One Long Sentence is All It Takes To Make LLMs Misbehave Feedly Summary: AI Summary and Description: Yes Summary: The text discusses a significant security research finding from Palo Alto Networks’ Unit 42 regarding vulnerabilities in large language models (LLMs). The researchers explored methods that allow users to bypass…
-
Schneier on Security: We Are Still Unable to Secure LLMs from Malicious Inputs
Source URL: https://www.schneier.com/blog/archives/2025/08/we-are-still-unable-to-secure-llms-from-malicious-inputs.html Source: Schneier on Security Title: We Are Still Unable to Secure LLMs from Malicious Inputs Feedly Summary: 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…
-
The Register: One long sentence is all it takes to make LLMs misbehave
Source URL: https://www.theregister.com/2025/08/26/breaking_llms_for_fun/ Source: The Register Title: One long sentence is all it takes to make LLMs misbehave Feedly Summary: Chatbots ignore their guardrails when your grammar sucks, researchers find Security researchers from Palo Alto Networks’ Unit 42 have discovered the key to getting large language model (LLM) chatbots to ignore their guardrails, and it’s…
-
Slashdot: Microsoft Reportedly Cuts China’s Early Access to Bug Disclosures, PoC Exploit Code
Source URL: https://it.slashdot.org/story/25/08/22/2059255/microsoft-reportedly-cuts-chinas-early-access-to-bug-disclosures-poc-exploit-code Source: Slashdot Title: Microsoft Reportedly Cuts China’s Early Access to Bug Disclosures, PoC Exploit Code Feedly Summary: AI Summary and Description: Yes Summary: Microsoft has modified its Microsoft Active Protections Program (MAPP) to restrict access to proof-of-concept exploit code for companies in certain countries, including China, to combat the leak of sensitive…
-
The Register: DeepSeek’s new V3.1 release points to potent new Chinese chips coming soon
Source URL: https://www.theregister.com/2025/08/22/deepseek_v31_chinese_chip_hints/ Source: The Register Title: DeepSeek’s new V3.1 release points to potent new Chinese chips coming soon Feedly Summary: Point release retuned with new FP8 datatype for better compatibility with homegrown silicon Chinese AI darling DeepSeek unveiled an update to its flagship large language model that the company claims is already optimized for…
-
The Register: Microsoft reportedly cuts China’s early access to bug disclosures, PoC exploit code
Source URL: https://www.theregister.com/2025/08/21/microsoft_cuts_chinas_early_access/ Source: The Register Title: Microsoft reportedly cuts China’s early access to bug disclosures, PoC exploit code Feedly Summary: Better late than never after SharePoint assault? Microsoft has reportedly stopped giving Chinese companies proof-of-concept exploit code for soon-to-be-disclosed vulnerabilities following last month’s SharePoint zero-day attacks, which appear to be related to a leak…