Abhinav Narne - Writing https://www.abhinavnarne.com/writing Articles on AI security, LLM red teaming, and trust & safety by Abhinav Narne. en-us Thu, 26 Feb 2026 02:58:01 GMT Notes on the Claude Code 2.1.0 outage https://www.abhinavnarne.com/writing/claude-code-outage A changelog formatting change took down Claude Code. Lessons about parsing human docs as machine data. Thu, 08 Jan 2026 12:00:00 GMT https://www.abhinavnarne.com/writing/claude-code-outage Notes on the Eurostar chatbot "vulnerability" report https://www.abhinavnarne.com/writing/eurostar-chatbot-analysis Looking at what makes something a vulnerability versus a hardening opportunity in LLM applications. Mon, 05 Jan 2026 12:00:00 GMT https://www.abhinavnarne.com/writing/eurostar-chatbot-analysis What I learned shipping 1,000+ PRs with Claude Code https://www.abhinavnarne.com/writing/shipping-with-claude-code Notes from using Claude Code in parallel git worktrees: Plan Mode, ultrathink, verification loops, and Chrome automation. Sat, 03 Jan 2026 12:00:00 GMT https://www.abhinavnarne.com/writing/shipping-with-claude-code How AI Regulation Changed in 2025 https://www.promptfoo.dev/blog/ai-regulation-2025/ Why "AI compliance questions" appeared in security questionnaires and RFPs, and how policy becomes contract requirements. Mon, 15 Dec 2025 12:00:00 GMT https://www.promptfoo.dev/blog/ai-regulation-2025/ Why Attack Success Rate (ASR) Isn't Comparable Across Jailbreak Papers https://www.promptfoo.dev/blog/asr-not-portable-metric/ ASR isn't portable across papers because measurement choices dominate the headline number. Includes math and a checklist for reading papers. Fri, 12 Dec 2025 12:00:00 GMT https://www.promptfoo.dev/blog/asr-not-portable-metric/ GPT-5.2 Initial Trust and Safety Assessment https://www.promptfoo.dev/blog/gpt-5.2-trust-safety-assessment/ Day-zero red team of GPT-5.2 focusing on jailbreak resilience and harmful content. Thu, 11 Dec 2025 12:00:00 GMT https://www.promptfoo.dev/blog/gpt-5.2-trust-safety-assessment/ Real-Time Fact Checking for LLM Outputs https://www.promptfoo.dev/blog/llm-search-rubric-assertions/ Introduces search-rubric, an assertion where a search-enabled judge verifies time-sensitive claims during evals and CI. Fri, 28 Nov 2025 12:00:00 GMT https://www.promptfoo.dev/blog/llm-search-rubric-assertions/ When AI becomes the attacker: The rise of AI-orchestrated cyberattacks https://www.promptfoo.dev/blog/anthropic-threat-intelligence-vibe-hacking/ Connects malware querying LLMs at runtime with "vibe hacking" case studies. Defense needs continuous testing. Mon, 10 Nov 2025 12:00:00 GMT https://www.promptfoo.dev/blog/anthropic-threat-intelligence-vibe-hacking/ Reinforcement Learning with Verifiable Rewards Makes Models Faster, Not Smarter https://www.promptfoo.dev/blog/rlvr-explained/ RLVR gains are often "search compression" rather than new reasoning ability. Fri, 24 Oct 2025 12:00:00 GMT https://www.promptfoo.dev/blog/rlvr-explained/ Prompt Injection vs Jailbreaking: What's the Difference? https://www.promptfoo.dev/blog/jailbreaking-vs-prompt-injection/ Jailbreaking targets model safety training; prompt injection targets application trust boundaries. Mon, 18 Aug 2025 12:00:00 GMT https://www.promptfoo.dev/blog/jailbreaking-vs-prompt-injection/ AI Safety vs AI Security in LLM Applications: What Teams Must Know https://www.promptfoo.dev/blog/ai-safety-vs-security/ Safety protects people from harmful outputs; security protects systems from adversarial manipulation. Sun, 17 Aug 2025 12:00:00 GMT https://www.promptfoo.dev/blog/ai-safety-vs-security/ Evaluating political bias in LLMs https://www.promptfoo.dev/blog/grok-4-political-bias/ Open methodology and dataset (2,500 political statements) to measure political leaning in models. Thu, 24 Jul 2025 12:00:00 GMT https://www.promptfoo.dev/blog/grok-4-political-bias/