Abstract Creative cognition involves linking weakly or unrelated concepts, enabled by semantic control (inhibiting dominant associations to retrieve weaker ones) or through spreading activation within the semantic system. Semantic aphasia (SA) patients have impaired semantic control despite relatively preserved semantic representations. To date, no studies have examined creativity in SA. It remains unclear how impaired control affects patients' creative potential, and whether spreading activation alone supports this. Creative potential was assessed across three experiments. Experiments 1 and 2 involved 11 SA patients and 25 controls; Experiment 3 included 13 SA patients and 14 controls. In Experiment 1 (category judgement), participants selected five targets from distractors across 24 categories with differing coherence levels (shared features among members). Experiment 2 (constrained category fluency) involved generating five exemplars per category. Creative potential was measured via uniqueness, flexibility, semantic distance and creativity ratings. Experiment 3 (unconstrained fluency) asked participants to name as many Animals as possible in 1 minute, with additional measures of clustering and switching. Although SA cases were unable to shape retrieval to pre‐defined associations (in the category judgement task), they showed creative potential in the constrained fluency task. In the unconstrained fluency task, patients were less able to use strategies. However, with fluency controlled, no group differences in creative potential existed. These findings provide the first neuropsychological evidence that spreading activation, even with impaired semantic control, can support creative responses. Creative potential in SA depends on task demands, aligning with broader findings of patients' sensitivity to context.