# "address poisoning" — X 热门讨论 (2026-09-24 21:37 UTC)

## @Rizvi_On_X (DeRizvi) · 09-24 20:24 · ♥29 ↻2 💬30 Good night family

I still see people losing money over 7 matching characters.

That’s address poisoning.

Someone drops a look-alike wallet into your history. You check the first 4 and last 4, paste it, and the funds end up somewhere else.

Send-to-Name™ removes that step.

You pay a FortressName, and the wallet generates a fresh one-time address for that payment. Only you and the recipient can derive it, while settlement stays on the native chain.

Checking hex isn’t a security model.

Claim a free FortressName: https://t.co/iUgdc74aIH

@Americanfort_io > 引用 @Rizvi_On_X: good night brothers.

A free mint is easy. The real question is what the NFT unlocks.

CyberThrone holders don’t just get a PFP. After verification, the HUB opens a digital identity pack tied to your robot:

high-res art, transparent PNG, avatar, mobile + desktop wallpapers, X/LinkedIn banners, a rigged 3D model, an activity book, and the Dawn of CyberThrone album.

The 3D file and album are original, non-AI work.

8,888 supply • FREE mint • Date TBA • Omnichain on Unvault

The JPEG is the key. The pack is the product. https://t.co/d1u2RDnSOP

@0xCyberThrone @unvault https://x.com/Rizvi_On_X/status/2103218920547385425

## @MastrXYZ (MASTR) · 09-24 21:01 · ♥36 ↻8 💬5 What is happening at @Bitget is unusual.

I have several findings that point to a potential problem, including substantial wallet outflows, conversions into ETH and an Ethereum withdrawal restriction reported by Bitget’s own API.

➡️The short version is in the 2nd post for anyone who does not want to read the full investigation.⬅️

I checked transaction records, token transfers, internal ETH movements, wallet labels and Bitget’s public API.

The snapshot below is from now September 24, 2026, approximately 20:15 - 20:45 UTC. A hack remains unconfirmed because the critical question is still unresolved: who controls the receiving wallets, and were these transfers authorised?

The primary collection address is:

0x770b10b273fc44fe9197d6bf20f145c2e98463ee

On Ethereum, I traced incoming transfers of 24,373.375940615 ETH, 34,751,168.120990 USDT, 12,852,046.242513 USDC and 3,000.322053 XAUt.

On Avalanche, the same address received 821,011.971142467 native AVAX.

At prices retrieved around the snapshot, those Ethereum and Avalanche inflows alone represent approximately $134.44M.

That is a verified subset of the movements.

The Ethereum sources are addresses independently labelled by explorers as Bitget 6, Bitget 35, also identified as Cold Wallet 9, and Bitget 5. Bitget 6 supplied approximately 9,010.90 ETH plus the USDT and USDC.

Bitget 35 supplied approximately 15,362.48 ETH. Bitget 5 supplied the XAUt. Explorer labels are attribution evidence, although they cannot prove who authorised a transaction.

The first recorded ETH funding of the collection address was 0.84 ETH at 18:31:11 UTC.

The large USDT transfer followed at 18:58:59. At 19:05:11, the collection address sent 0.1 ETH to a second wallet, followed by practically all the USDT, USDC and XAUt:

0x7c96279ec1e888aa56b9b836e0db26ca48573e1c

That second wallet started converting the assets into ETH. This is supported by token outflows and native ETH receipts sharing transaction hashes. For example, this transaction returned approximately 1,855.7977 ETH:

https://t.co/Pl1ENBSzr9

At 19:30:35 UTC, the swap wallet forwarded 22,320 ETH to:

0xa6dd3f218b65e32ccc37be30f74884133c655545

https://t.co/7q2ou2cWZf

The Bitget-labelled source wallets continued sending ETH to the collection address after those conversions had begun. At 20:09:11 and 20:09:23 UTC, another 1,879.2 ETH and 1,395.9 ETH arrived.

The collection address then distributed 20,000 ETH through 2 separate transfers: 10,000 ETH at 20:13:11 UTC to 0xd2c2f029…2d9f899, followed by 10,000 ETH at 20:19:11 UTC to 0x600cfedc…45784b2.

https://t.co/zy9Jr35oDn

https://t.co/wCnXvCWC9X

Those downstream transfers are movements of assets already counted. Adding them again would inflate the total. The collection address still held approximately 4,373.28 ETH and, separately on Avalanche, 821,011.97 AVAX at the snapshot.

🚨 Another concrete finding comes directly from Bitget.

Its public coin-status API returned withdrawable=false for ETH on the Ethereum network, while rechargeable remained true. Several other withdrawal routes, including native BTC, BNB on BEP20 and AVAX C-Chain, remained marked withdrawable=true.

https://t.co/deySwBMXeQ

The API itself was accessible. A claim that all withdrawals or APIs were shut down would go beyond the evidence.

There is also contamination in the transaction history.

Fake tokens using the names USDT, USDC and ETH imitate genuine transfers, sometimes copying their amounts.

One fake USDT contract, 0x34786b43b696f34de244fb76730c79b876eb9177, reproduces a 34,751,168.12 USDT movement.

It is not the actual USDT contract. I excluded these entries.

Tiny transfers also originate from addresses copying the beginning and end of legitimate counterparties, consistent with address poisoning.

An unsolicited 0.069 RAIL transfer appears too. Neither that token transfer nor the poisoning activity establishes who controls the collection wallet, and the RAIL transfer is not evidence that these funds entered Railgun.

The main transaction dataset came from Routescan, with 1 substantial ETH transfer additionally corroborated through Ethereum JSON-RPC.

The official maintenance notices I found concern Injective. They do not explain these Ethereum movements.

Multiple exchange-labelled wallets feeding a collection address, assets being converted into ETH, onward distribution and a withdrawal restriction together warrant a precise explanation.

Bitget needs to clarify whether it controls these destinations and authorised the conversions.

Until that is established, calling this routine treasury management would be premature, and publishing “$183M confirmed stolen” would also exceed the evidence. https://x.com/MastrXYZ/status/2103228316127813903

## @Svrkee01 (Ali) · 09-24 19:53 · ♥24 ↻1 💬29 GN Fam❤️

Crypto payments still expose more than they need to.

A public address can become a permanent mailbox, carrying its balance and payment history with it. Even a readable name can create the same problem if it always points to one fixed wallet.

@Americanfort_io takes a different approach with Send to Name™, which is already live.

With a FortressName, you pay an @name instead of copying a long address. The wallet generates a fresh one time address for that payment. The sender and recipient can derive it, while only the recipient can spend the funds. Settlement still happens on the native chain.

That means the name can stay public without becoming a public index of your balance or payment history.

It also removes the copy paste surface behind address poisoning attacks. You pay the name instead of checking hex characters.

One FortressName currently works across 13 integrated networks and is designed as infrastructure for existing wallets, not another destination app.

SafeSend™ is still in development, so the live product today is Send to Name™.

The recipient still knows who paid them. The privacy is about keeping the public payment trail from becoming a permanent lookup.

Free FortressName: https://t.co/NmQpBXBKtv

@Americanfort_io https://x.com/Svrkee01/status/2103211293427777602