It’s not. I’ve had a similar experience, and I walked away.
Instagram is everything poisonous about Facebook’s values and platform and practices, distilled. They get nothing from me except a temporary email address, a proxy IP, and a false user-agent string.
I'm going to gander a wild guess and say the combination of temporary email address, proxy IP, and false UA string are likely to trip up fake account detection ML algorithms.
The UA string is real, just misleading. The other two resources aren’t public throwaways, the domain and source addresses belong to me, or a version of me (as much as IANA allocated resources can “belong”) and otherwise appear normal, but are only used for traffic to untrustworthy shitweasels.
It’s not impossible, but it is unlikely that it looks like a bot.
I figured it was actually the other way around, or at least a slightly different shade of things; not so much that it looked like a fake account, more than it looked like someone they couldn’t correlate to any other identity and therefore has no consumer surveillance value, ergo I can either fill in the blanks, or fuck off. Whatever the truth of it, I chose the latter
Instagram is everything poisonous about Facebook’s values and platform and practices, distilled. They get nothing from me except a temporary email address, a proxy IP, and a false user-agent string.