Server-Side Bug Allowing Malicious Actors in Voice Chats

[IRGC]

Hello @levlam ,
I’m reporting a potential server-side bug affecting our large Telegram supergroups, where users engage in text and voice chats. We’re frequently disrupted by malicious actors who join voice chats as private channels and verbally abuse admins and participants. These individuals were previously banned for inappropriate behavior, yet they persist in joining.
When we try to remove these abusive private channels from the voice chat participant list, they remain despite the mobile app showing them as banned. Attempts to address this via userbot or MTProto result in a CHANNEL_PRIVATE error.
This issue has been ongoing for a long time. I’ve submitted multiple reports via the TDLib bot and bugs.telegram.org, but I’ve received no response. I’m now seeking your help.
Please provide clear answers to these questions:
Are private channels meant to join voice chats by design?

If yes, why can’t we ban or kick them?

If no, how are they able to join voice chats?

I’d appreciate your prompt attention to this matter.
Thank you.