19cbaecb-5

ConfirmedSecurity589 words3 min decode

Bitget Wallet Flags 37,000 Weekly Phishing Signing Requests

Bitget Wallet flagged about 37,000 phishing-related signing requests weekly in Q3 2026, 7.4 times Q2’s average, as it expanded AI-assisted GetShield detection.

Bitget Wallet Phishing Alerts Climb Sevenfold to 37,000 a Week
WitnessBitget Wallet Phishing Alerts Climb Sevenfold to 37,000 a WeekFoundry / pixabay

Outputs

  1. Q3 2026 alerts averaged about 37,000 flagged signing requests a week, versus 5,000 in Q2.

  2. The Q3 weekly average was 7.4 times Q2’s and counted flagged interactions, not unique attacks or confirmed losses.

  3. Phishing-related signing requests represented 57% of Q3 detections, followed by malicious addresses at 17%, risky contracts at 14% and malicious approvals at 12%.

  4. AI identified 90% of newly blocked risky domains in October, up from 36% in Q2, while nearly 70% had no match in external feeds.

  5. A trader lost nearly $1 million in July after signing a malicious Ethereum token approval.

By Marcus Bennett

Bitget Wallet’s security engine averaged about 37,000 flagged signing requests a week in Q3 2026, 7.4 times the 5,000 weekly average in Q2, the wallet said Nov. 11 as it expanded AI-assisted threat detection.

The headline figure needs context. The alerts count flagged signing interactions, not unique attacks or confirmed financial losses. Better detection drove part of the increase, alongside a potentially more active threat environment.

Alvin Kan, Bitget Wallet’s chief operating officer, shared the figures with Cointelegraph. He said: “The surge likely reflects both a more active threat environment and sharper monitoring.” The initial AI-assisted risk system went live in Q2, so it had operated for less than two full quarters when the figures were compiled.

GetShield’s wider checks

The Nov. 11 announcement expanded GetShield across website, domain and transaction analysis. GetShield scans website code for fraud or impersonation. It checks domain registration records and connections to known malicious domains. The engine also combines its own screening with external threat intelligence.

Before a user signs, GetShield simulates the transaction’s outcome. It looks for hidden permissions and gaps between what the screen displays and what the signature authorizes. A crafted approval can give an attacker access to funds without exposing the private key.

The system manages token approvals across websites, decentralized applications and smart contracts. It also looks for unexpected last-second changes to signing requests.

Phishing-related signing requests accounted for 57% of Q3 detections across four monitored categories:

  • Phishing-related signing requests: 57%
  • Malicious addresses: 17%
  • Risky contracts: 14%
  • Malicious approvals: 12%

In October, AI identified 90% of newly blocked risky domains, up from 36% in Q2. Nearly 70% of those flagged domains had no matching alerts in any external threat feed Bitget Wallet checked. That blind spot was the company’s core reason for moving beyond third-party databases. Its own screening had found risks the outside lists missed.

What does this change for wallet users?

The practical answer is earlier intervention before a signature grants broad control over funds. In July, a trader lost nearly $1 million after signing a malicious Ethereum token approval. One signature created the path to drain the wallet, without a compromised seed phrase. GetShield’s simulation looks for that gap between an ordinary approval and its actual authority.

In comments published Oct. 6 and reported by crypto.news, Kan argued that wallets need to state plainly whether an approval covers one payment or lets an app move funds indefinitely. He also argued that revoking existing permissions should be straightforward. Bitget Wallet frames the broader goal as catching risks earlier and limiting damage while users retain control.

As self-custodial wallets move deeper into payments and day-to-day finance, the distinction can determine whether a signature is routine or an open-ended spending mandate. The expansion adds more checks before that decision while leaving final approval with users.

The next operational test

Bitget Wallet plans to widen AI-driven threat detection and explore AI-assisted incident response. The proposed response layer would identify emerging risks and accelerate action, with humans staying in the loop. The company said Nov. 11 that the broader program now covers website, domain and transaction analysis. Its next test is operational performance as coverage expands.

No deadline for the incident-response work was disclosed. As self-custodial wallets push deeper into payments and everyday finance, Bitget Wallet will need to show whether earlier detection reduces harmful approvals without taking signing decisions away from users.

via bitcoinfoundation.org (Source)

More from Marcus Bennett

Marcus Bennett

Show full bio

Senior reporter covering money, funding and market shifts around crypto and web3 industry at Mempool Brief.

1,581 articles