prompt injections
PulseAugur coverage of prompt injections — every cluster mentioning prompt injections across labs, papers, and developer communities, ranked by signal.
1 day(s) with sentiment data
-
New research details LLM "role spoofing" vulnerability and mitigation
A new research paper explores the vulnerability of large language models (LLMs) to "role spoofing," a type of prompt injection where malicious text can impersonate a user by mimicking their writing style or formatting. …
-
Huawei AI chips challenge Nvidia, Meta pushes enterprise AI, AMD launches Gorgon Halo · 1 source tracked
Several AI-related developments are emerging across the tech landscape. Huawei's CEO claims their homegrown AI chips are outselling Nvidia in China, despite potential limitations in advancement. Meta is pushing enterpri…
-
New benchmark reveals AI agents vulnerable to harmful actions
Researchers have developed a new benchmark called BAD-ACTS to evaluate the robustness of AI agentic systems against adversarial attacks designed to elicit harmful actions. The benchmark includes five distinct agentic sy…
-
Job candidates use AI prompt injections to bypass application screening
Job candidates are increasingly using AI prompt injections to bypass automated screening systems and get their applications noticed. This tactic involves embedding hidden instructions within resumes or cover letters tha…