How does Discord alt detection work?
A bot sees an ID and a join time. A verification page sees much more. What alt detectors measure, and why they get it wrong.
| Account age | 2 days |
|---|---|
| VPN or proxy | Yes |
| Network (ASN) | Hosting provider |
| Captcha | Passed |
“Alt detector” covers two very different things. One is a bot in your server that looks at new accounts. The other is a verification step that members go through in a browser before they get access. They see different data, so they catch different alts. Here is what each can actually see, and why neither one can promise certainty.
What a bot in your server can see
When someone joins, Discord tells every bot in the server roughly what you can see yourself: user ID, username, display name, avatar, public badges and the moment they joined. The ID matters most, because Discord IDs encode their own creation time. Discord’s developer docs define it as milliseconds since the Discord Epoch (the first second of 2015) in the top 42 bits.
So a bot can say “this account is two hours old” with confidence. It cannot see an email, a phone number or an IP address. Discord does not hand those to bots. A bot-only detector is really an account-age and profile checker.
What a verification page can see
A verification step sends the member to a web page, usually after they approve the app on Discord’s own authorization screen. Two things change:
- The page receives a normal web connection, so it sees the IP address, the browser and roughly where the connection comes from.
- With the member’s permission, Discord’s OAuth2 can share more account details. The “email” scope, for example, lets an app read the account’s email address.
Now the detector can compare. Has this IP verified before on an account you banned? Is the connection a VPN, a proxy or a data-centre network (identified by its ASN, the number that names a network operator)? Is it coming from a country you block? That is real alt detection, and it is also where the mistakes come from.
| Signal | Catches | False positive risk |
|---|---|---|
| Account age | Throwaway accounts made after a ban | Low, but real newcomers are new too |
| Same IP as a banned account | Evaders on the same home connection | Medium: households, dorms, schools |
| VPN or proxy | Evaders hiding their home IP | Medium: privacy-minded members |
| Cellular network | Evaders hopping mobile IPs | High: mobile IPs are shared widely |
| ASN (hosting networks) | Cheap servers and proxy farms | Low for hosting, higher for small ISPs |
| Country | Raids from a region you do not serve | Depends on your audience |
| Known scammer database | Accounts already flagged elsewhere | Low, but only covers reported accounts |
Why no detector is certain
Every signal above describes a connection or an account, not a person. Two people can share one connection. One person can use two connections. A determined evader who waits a few weeks, uses a different network and changes their writing style can look like anyone else. Any tool that promises it “finds every alt” is overselling.
The honest way to use detection is as a filter with a human behind it: block the obvious cases (an account made ten minutes ago, on a hosting network, matching a banned IP) and send the borderline ones to a mod. For the manual side, the guide on how to check for alts on Discord covers profile and behaviour tells.
Settings that cause the most false flags
- Blocking cellular networks. It stops IP hopping, and it also blocks people verifying on mobile data.
- Very long minimum account ages. A 90-day minimum stops throwaways and every real newcomer at once.
- Automatic bans on an IP match. Send matches to review instead, at least until you trust the pattern.
Is it allowed?
Running checks on people who choose to verify in your server is normal moderation. What you do with the data matters: use it to decide who gets in, keep it inside your staff team and never publish someone’s details. For what Discord itself says about alts, see does Discord allow alt accounts.
Questions people ask
Do Discord alt detector bots work?
Partly. A bot that only sits in your server sees account age, profile and join time, so it can catch fresh throwaway accounts and little else. Detectors that send members through a verification web page can also compare IP addresses and network details, which catches far more, with some false positives.
Can an alt detector be wrong?
Yes, in both directions. Siblings, roommates, a school or a mobile network can share an IP, so innocent people get flagged. A careful evader with an aged account and a new connection can pass. Good tools flag for review instead of banning automatically.
Is there a Discord alt detector website I can search?
No public site can search the data needed, because Discord does not share emails, phone numbers or IPs with anyone. Detection only works on the people who pass through your own verification step.
Does a VPN beat alt detection?
A VPN hides the home IP, but VPN and hosting networks are themselves detectable. That is why detectors check for VPNs, proxies and the network (ASN) a connection comes from, not just the IP.