"The extant research shows that for most people with ASD [autism spectrum disorder] there is no association between ASD and delinquent behavior."
That was one of the primary conclusions reached in the literature review published by Alexa Rutten and colleagues  (open-access) looking at the collected peer-reviewed science from 1990 to 2015 on this topic. Boiling down the literature to 12 papers "five of which report the prevalence of delinquency in patients with ASD and seven the prevalence of ASD in a forensic population", authors reported on quite a lot of variability when it came to delinquency and offending behaviours but overall, offending behaviour was "lower in people with ASD than in the general population."
This is welcomed research. It reiterates what many people have already known/suspected, in that the label 'autism' for many is not typically associated with intentional delinquency or offending behaviour. As Rutten et al note: "many people with ASD have an overactive sense of right and wrong and are usually conscientious and unwilling to break the law." I should also point out that a diagnosis of autism is not however necessarily some kind of 'shield' when it comes to offending behaviour (see here and see here for a more recent example) or indeed, potential 'vulnerability' to becoming involved in specific offences (see here). But serious 'intentional' delinquency is not the norm; even that is, in the context of something like greater likelihood of contact with law enforcement agencies (see here).
There are still lessons to learn in this area of research and practice. Certain 'over-represented' comorbidity appearing alongside autism is still something to potentially consider  as per other findings (see here and see here). I say that without 'trying to pass the [diagnostic] buck'. Bearing in mind also how wide the autism spectrum is, specific diagnoses on the spectrum might also require further study as per the authors comments: "The prevalence of ASD diagnoses, particularly Asperger’s syndrome, in forensic settings is remarkable because it is much higher than the prevalence of ASD diagnoses in the general population." At this point I'll also refer you once again to the comprehensive review paper by Tom Berney  that mentions some of the potential how-and-whys of offending in the specific context of Asperger syndrome, again minus any sweeping generalisations. Recent media attention on previously undiagnosed Asperger syndrome in a prison context also makes for important and relevant reading too (see here) particularly where substance abuse is prominently mentioned (see here).
I would however question one rather sweeping statement made by Rutten and colleagues: "It is important to diagnose ASD carefully and to differentiate autism symptoms such as a lack of empathy from psychopathic traits". Lack of empathy and autism? Hmm, sounds a bit old hat to me...
I will again close with the point made that for the majority of people on the autism spectrum, the label is much more likely to be associated with law abiding rather than law breaking.
 Rutten AX. et al. Autism in adult and juvenile delinquents: a literature review. Child and Adolescent Psychiatry and Mental Health. 2017; 11: 45.
 Newman SS. & Ghaziuddin M. Violent crime in Asperger syndrome: the role of psychiatric comorbidity. J Autism Dev Disord. 2008 Nov;38(10):1848-52.
 Berney T. Asperger syndrome from childhood into adulthood. Advances in Psychiatric Treatment. 2004. 10; 341-351.