/en/tags/cybersecurity/Recent content in Cybersecurity on The Engine RoomHugo -- 0.155.3en-usTue, 26 May 2026 12:00:00 +0300/en/notes/ai_cybersecurity_careers_2026/Tue, 26 May 2026 12:00:00 +0300/en/notes/ai_cybersecurity_careers_2026/Career overview of the AI Security market for 2026/en/notes/ai_security_-courses/Thu, 16 Apr 2026 15:00:00 +0300/en/notes/ai_security_-courses/Courses and certifications in AI Security/en/notes/mcpthreathive/Thu, 16 Apr 2026 00:00:00 +0300/en/notes/mcpthreathive/A threat discovery platform for Model Context Protocol/en/notes/rlspoofer/Tue, 14 Apr 2026 15:00:00 +0300/en/notes/rlspoofer/An attack on watermarks as a way to spoof authenticity/en/notes/deepsight/Sun, 15 Feb 2026 15:00:00 +0300/en/notes/deepsight/A transition from black-box evaluation to transparent AI safety diagnostics/en/notes/ieee_tech_predictions_2026/Mon, 02 Feb 2026 15:00:00 +0300/en/notes/ieee_tech_predictions_2026/IEEE released a report covering the key technology trends expected to shape 2026./en/notes/open_rt/Tue, 23 Dec 2025 15:00:00 +0300/en/notes/open_rt/OpenRT is a modular and extensible environment for systematic safety evaluation of large language models/en/notes/slm/Mon, 15 Dec 2025 15:00:00 +0300/en/notes/slm/Note on the document Small Language Model for AI Agents HandBook/en/notes/doublespeak/Wed, 10 Dec 2025 15:00:00 +0300/en/notes/doublespeak/The authors present a new attack called Doublespeak: a simple attack based on hijacking the model’s internal representations in context/en/notes/finesec/Sat, 06 Dec 2025 15:00:00 +0300/en/notes/finesec/A new framework for creating compact models for finding vulnerabilities in C/C++ code/en/notes/whisper_leak/Thu, 04 Dec 2025 15:00:00 +0300/en/notes/whisper_leak/A new attack that makes it possible to determine the topic of an LLM query from encrypted traffic/en/notes/breaking_agent_backbones/Tue, 02 Dec 2025 15:00:00 +0300/en/notes/breaking_agent_backbones/How LLM selection affects agent security/en/notes/lotl_attack_with_llm/Sun, 30 Nov 2025 15:00:00 +0300/en/notes/lotl_attack_with_llm/How future devices with built-in LLMs will become a security problem, because attackers will be able to live off the LLM (LOLLM)/en/notes/architecting_secure_enterprise/Tue, 25 Nov 2025 15:00:00 +0300/en/notes/architecting_secure_enterprise/A guide to designing secure enterprise AI agents using MCP from IBM, with verification from Anthropic/en/notes/defence_mllm_from_jailbreak/Sat, 22 Nov 2025 15:00:00 +0300/en/notes/defence_mllm_from_jailbreak/A new class of attacks where text and image look safe separately, but their combination carries malicious meaning/en/notes/pruning_activated_attack/Mon, 17 Nov 2025 15:00:00 +0300/en/notes/pruning_activated_attack/Model pruning can be used by an attacker/en/notes/data_extraction_attacks_against_rag/Fri, 14 Nov 2025 15:00:00 +0300/en/notes/data_extraction_attacks_against_rag/The paper studies a new class of attacks against RAG-type systems/en/notes/fine_tuning_jailbreaks/Mon, 10 Nov 2025 15:00:00 +0300/en/notes/fine_tuning_jailbreaks/The paper discusses vulnerabilities in fine-tuning systems for large language models under conditions close to real-world operation/en/notes/airgeddon_wifi/Fri, 07 Nov 2025 15:00:00 +0300/en/notes/airgeddon_wifi/Nobody likes wires; everyone loves Wi-Fi/en/notes/tool_tweak/Thu, 06 Nov 2025 15:00:00 +0300/en/notes/tool_tweak/An attack on tool selection in agentic systems/en/notes/func_poison/Mon, 03 Nov 2025 15:00:00 +0300/en/notes/func_poison/A new attack based on poisoning a function library, which makes it possible to replace agent behavior without changing their models/en/notes/red_teaming_llm_with_mcp/Thu, 30 Oct 2025 15:00:00 +0300/en/notes/red_teaming_llm_with_mcp/MCP is a critical and vulnerable point in the trust chain of LLM agents/en/notes/fuzzingbrain/Thu, 23 Oct 2025 15:00:00 +0300/en/notes/fuzzingbrain/All you need is fuzzing…/en/notes/kerberos/Tue, 21 Oct 2025 15:00:00 +0300/en/notes/kerberos/A protocol that allows users to authenticate on a network and access services after authentication/en/notes/ntlm/Mon, 20 Oct 2025 15:00:00 +0300/en/notes/ntlm/NTLM is a Microsoft authentication protocol/en/notes/shell_or_nothing/Sun, 19 Oct 2025 15:00:00 +0300/en/notes/shell_or_nothing/An article about an LLM-based framework and its results in obtaining shell access/en/notes/bit_flip_attacks/Fri, 17 Oct 2025 15:00:00 +0300/en/notes/bit_flip_attacks/The article describes the problem of understudied hardware attacks on LLMs: bit-flip attacks (BFA)/en/notes/cuckoo_attack/Wed, 15 Oct 2025 15:00:00 +0300/en/notes/cuckoo_attack/An analysis of a new attack against IDEs with LLM agents: how AI assistants can become a tool for stealthy attacker persistence in a system./en/notes/multiagent_pipeline/Wed, 15 Oct 2025 15:00:00 +0300/en/notes/multiagent_pipeline/An analysis of a multi-agent defense architecture that reduces prompt injection attack success by separating roles between agents.