https://silan.tech/blog/ Writing on AI systems, data, runtimes, and dependable execution. en https://silan.tech/blog/gem-bench-ai-answers-with-ads/ https://silan.tech/blog/gem-bench-ai-answers-with-ads/ GEM-Bench turns one practical failure into a repeatable test: after a paid recommendation enters an AI answer, does the answer remain accurate, natural, and trustworthy, and what does that quality cost? benchmarks gem-bench generative-advertising kdd-2026 llm-evaluation https://silan.tech/blog/make-research-work-findable/ https://silan.tech/blog/make-research-work-findable/ Make research work findable with a local-first workflow that keeps claims, evidence, project pages, researcher profiles, and AI/search metadata consistent. ai-readable-research make-research-work-findable research-discoverability research-publishing silan-viking