Ample evidence shows that post-encoding misinformation from others can induce false memories. Here, we demonstrate in two experiments a new, tacit form of socially generated false memories, resulting from interpersonal co-monitoring at encoding without communication of misinformation. Pairs of participants jointly viewed semantically coherent word lists, presented successively in blue, green, or red letters. Each individual was instructed to memorize words presented in one of the colors. One color remained unassigned (control condition). Participants (total N = 113) reported more false memories for non-presented words (lures) semantically related to partner-assigned than to control lists, although both list types were equally irrelevant to their own task. Notably, this effect also persisted for particularly rich memories. These findings show for the first time that social induction of false memories, even subjectively rich ones, does not necessarily require communication of deceptive information. This has important implications both theoretically and practically (e.g., in forensic contexts).