Struggling to distinguish mere preference from true teaching-away in AI prior art—or to prove unexpected results with legal weight? By the end of this lesson, you’ll diagnose dissuasion language with precision, build secondary-consideration records that tie claim elements to metrics, and draft examiner-ready phrasing that survives §§102/103. You’ll get crisp explanations, corpus-driven templates, high-signal examples, and targeted exercises to validate your judgment and tighten your claim charts. The approach is discreet and surgical—focused on measurable ROI: fewer OAs, stronger nonobviousness, faster allowances.