Academic audits likely underestimate AI-hallucinated citations in preprints because they only check titles in online databases, missing errors in author attribution, journal placement, and non-journal sources. Analysis of 46 recent hallucinations shows nearly half involve real papers with wrong authors or journals, suggesting current prevalence estimates may need to be doubled.
Minsoo Lee, a 24-year game designer and former lead on Ragnarok Online, published a free field manual on GitHub about integrating Claude Code into game production workflows. The book covers practical applications from systems design to live ops, emphasizing validation gates to prevent AI hallucinations, schema-first data design, and cost management, with real examples from shipping an MMORPG in Korea.
Spanda is a Rust-based tool that detects LLM hallucinations and quantifies epistemic uncertainty in sub-microsecond time without requiring expensive neural cross-encoders. It uses Exact-Match Normalized Entropy to match or exceed traditional Semantic Entropy methods while operating ~90,000× faster, making it practical for high-throughput production serving.
A VerusCite tool developer criticizes Springer Nature's AI citation-checking system, arguing it fails to catch hallucinated references in published articles from August–September 2026. The author provides examples across multiple major publishers and argues that for $2 per paper, publishers should invest more in reference verification rather than relying solely on peer review.