This study explores how Indian faculty are warming up to AI teaching assistants amid packed classrooms and NEP 2020 pressures, blending surveys (195 responses) and interviews across IITs to state colleges. Findings reveal 58% acceptance driven by time savings (usefulness score: 3.8/5), but ease-of-use gaps (3.2/5) and ethics worries stall full adaptation—STEM youth lead, humanities lag. Hybrid models promise workload relief and better engagement if infra and bias issues get fixed. Implications urge AICTE/UGC for targeted training to evolve teaching without losing the human spark.