The only way I can see this as being possible
Sometimes it is rather difficult to reconcile data differences without input from the manufacturer or the organisation which undertook a post-manufacture STC activitiy.
Unfortunately the AFM is not the ultimate defense.
This view probably would need an appropriate legal opinion to support it. Note CAR (1988) 138 (1) as being the minimum standard. My (non-legally competent) take on this is that the AFM would provide a defence.
Certainly, if an operator/pilot held the view that details in the AFM were non-compliant with the certification basis, then that ought to be drawn to the attention of the Regulator in an appropriate manner. Either the operation should be suspended by the operator or, if the operator holds adequate technical competence, the operation should be varied conservatively with respect to the AFM in order to meet the operator's assessment of the certification requirement pending Regulatory response and possible action.
If a pilot believes the performance data is optimistic garbage
As a general consideration, I see AFM performance data for larger aircraft as being reasonably compliant with the design standards.
One needs to keep in mind that, at times, the certification data may appear to be a little artificial when viewed from out there in real world conditions. That perceived delta then comes within the ambit of an operator's corporate risk management protocols. I noted earlier that, at the now long defunct IPEC Aviation, we made such a decision in respect of first segment concerns for the Argosy. It is always open for any operator to operate to a higher than minimum standard.
At the end of the day, we need to consider two requirements - first, making sure we do not fall below extant regulatory requirements, and second, making sure that we don't scratch the paintwork.
Certainly the AFM is not always infallible. Over the years, I have picked up a couple of AFM technical errors (as well as quite a few administrative errors - including one helicopter which had had the wrong AFM since its initial CofA, years before) which, following subsequent Regulator review, were subject to amendment. Indeed, for a while there, when I was actively involved in Industry audit work, the Regulator was quite cheerfully happy to use me as a defacto AFM review and checking resource.