MediaScout Weekly

Sign in Subscribe

Latest

AI Agent Reliability: 74% of the Fix Is Boring

AI Agent Reliability: 74% of the Fix Is Boring

A February 2026 arXiv paper titled "Towards a Science of AI Agent Reliability" proposes twelve metrics for judging whether an autonomous agent deserves your trust, split across consistency, robustness, predictability, and safety (arXiv). But the number that should actually change how you build comes from a different study:

Jev: The Language-Free System One Model at $0.042

Jev: The Language-Free System One Model at $0.042

Adaptive Evidence Routing Turns Context Selection Into a Decision

Adaptive Evidence Routing Turns Context Selection Into a Decision

Open-Source Multimodal Embeddings: Google's 740M Model Rewrites the Math

Open-Source Multimodal Embeddings: Google's 740M Model Rewrites the Math

arXiv Submission Limit: Two Papers a Month, No Exceptions

arXiv Submission Limit: Two Papers a Month, No Exceptions

Query-Aware Multimodal Memory: Six Buckets Beat One Pile

Query-Aware Multimodal Memory: Six Buckets Beat One Pile

Open-Source Web Search for AI Agents Now Costs $0

Open-Source Web Search for AI Agents Now Costs $0