It would demonstrate that you punish behavioral signifiers like regularly interacting with people who disagree, and appearing on mass blocklists, and claiming that these are politically neutral measures.
The people who get caught in the broad net include anyone who goes against the mainstream progressive narrative.
You have data regarding how many times someone gets blocked. How many reports people are implicated in. How many "negative interactions" someone engages in.
Look at who is most likely to report others, and whether this is based on behavior or politically aligned content.
Who is more likely to interact with people they disagree with, and who is more likely to insulate themselves and block people at the drop of a hat?
You're either lying, or you're too incompetent to comprehend the full effects of the system that you engineered.
"Ok, but do you kill people?"
"We can categorically say that we do not murder people."