Background: Oxytocin administration, which may be of therapeutic value for individuals with social difficulties, is likely to affect endogenous levels of other socially relevant hormones. However, to date, the effects of oxytocin administration on endogenous hormones have only been examined in neurotypical individuals. The need to consider multi-hormone interactions is particularly warranted in oxytocin trials for autism due to evidence of irregularities in both oxytocin and sex steroid systems. Methods: In this double-blind cross-over study, saliva samples were collected from 16 autistic and 29 neurotypical women before and after intranasal administration of 24 IU oxytocin or placebo. Oestradiol, testosterone, and oxytocin levels were quantified in saliva samples. Participants also completed the Autism-Spectrum Quotient (AQ) and Empathy Quotient (EQ) questionnaires. Results: Distinct patterns of change in testosterone and oestradiol levels pre-to-post-administration were observed in autistic relative to neurotypical women (ANCOVA, p < 0.05 main effect of Group), controlling for sample collection time. The mean percent change oestradiol was + 8.8% for the autism group and − 13.0% for the neurotypical group (t = 1.81, p = 0.08), while the mean percent change testosterone was + 1.1% in the autism group and − 12.6% in the neurotypical group (t = 1.26, p = 0.22). In the oxytocin condition, the mean percent change oestradiol was + 12.6% in the autism group and − 6.9% in the neurotypical group (t = 1.78, p = 0.08), while the mean percent change testosterone was + 14.4% in the autism group and − 15.2% in the neurotypical group (t = 3.00, p = 0.006). Robust regression confirmed that group differences in percent change hormone levels were not driven by a small number of influential individuals. Baseline hormone levels did not differ between groups when considered individually. However, baseline testosterone relative to oestradiol (T:E2 ratio) was higher in autistic women (p = 0.023, Cohen's d = 0.63), and this ratio correlated positively and negatively with AQ and EQ scores, respectively, in the combined sample. Limitations: Further studies with larger and more diverse autistic sample are warranted to confirm these effects.