[Constitutional] AIP: Disable Legacy Tether Bridge

Proposals
1

Constitutional

Note: this proposal will be subject to an individual Snapshot vote; if it reaches a quorum of 3% of votable $ARB, it may be combined with several other maintenance-related proposals for a joint Tally vote.

Abstract

This proposal seeks to disable the legacy USDT bridge for Arbitrum One, given the activation of the new USDT0 bridge in January 2025. While the majority of crosschain volume has migrated to the new USDT0 bridge, the legacy USDT bridge is still technically enabled. This proposal seeks to disable the legacy bridge entirely, rather than continue with the current configuration which subjects depositors to a week-long delay and risks loss of funds if the depositor is a smart contract.

Motivation

Historically, the existing USDT bridge for Arbitrum One was the primary path for bridging Tether to and from Arbitrum One. The new USDT0 standard introduced a modernized alternative that has become the go-to solution. However, the legacy USDT bridge is still technically enabled, which presents several issues.

First, withdrawals from the legacy bridge to L1 are subject to a seven-day delay, which is meaningfully slower than the new USDT0 bridge which can process bridge transactions in seconds or minutes. Also, once this delay has elapsed, deposits are auto-withdrawn on L1. This inconvenience can be addressed by disabling the legacy bridge and directing users exclusively to the new USDT0 bridge.

Second, user funds could be lost if a smart contract deposits to the legacy bridge on their behalf. The legacy bridge is currently configured to auto-withdraw funds as soon as possible back to the depositor; however, this auto-withdraw configuration can cause problems for smart contracts that deposit on behalf of end users but do not have withdrawal or refund functionality. While it’s possible to recover funds in this scenario, it is certainly sub-optimal for smart contract operators and end users that delegate to them. Disabling the legacy bridge prevents smart contracts from encountering this issue and putting funds at risk. Also, once the legacy bridge is disabled, any transaction submitted to it will simply revert; this applies to third party integrations as well as direct user interactions.

Key Terms

Specifications

Deactivate Legacy Bridge: execute the ‘perform’ function on the DisableGatewayAction contract, disabling legacy USDT on the L1GatewayRouter contract

Steps to Implement

  1. Contract Execution: call DisableGatewayAction contract to disable legacy USDT
    • An audit report for the DisableGatewayAction contract can be found here

If this proposal passes a Snapshot vote successfully, it will be combined with other maintenance-related proposals for a unified Tally vote and on-chain payload. The portion of the Tally payload that covers the actions described in this proposal will be shared as soon as each individual proposal passes a Snapshot vote.

Timeline

The following timeline outlines proposed milestones from initial discussions to full implementation. Adjustments may be made based on community feedback and DAO governance requirements.

  1. June 29th: AIP forum post outlining the proposed discontinuation of the legacy USDT bridge for Arbitrum One (requisite code has already been written, tested, and audited as of 12 March 2025)
  2. July: Temperature check held on Snapshot
  3. July/August: Should the temperature check pass, as specified above, it will be included in the AIP draft combining several other maintenance-related proposals for a joint Tally vote
  4. On-chain Tally vote: Will commence once all other maintenance-related proposals have passed the respective temperature checks on Snapshot
  5. Post-on-chain Tally vote: Should the Tally vote pass, the waiting period on L2 will last 1 week after Tally vote is finalized, after which, the legacy bridge will be disabled
  6. Perpetuity: post-deprecation, any transaction submitted to the legacy bridge will revert
6 Likes
2

I support this proposal - migrating to the new bridge standard is a step in the right direction for ecosystem development.

However, I have a few questions:

  1. If the USDT0 bridge already exists and was implemented without a DAO vote, solely through an agreement between Tether and Offchain Labs, then why is this current proposal being brought to a vote?
    Literally this raises two questions:

    • why is this specific action being put to a vote now, and

    • why wasn’t the earlier migration to USDT0 subject to DAO approval?
      (For comparison, a similar bridge migration for USDS in the SKY project is also being handled via a DAO vote)

  2. I’d also like to ask: will there be any broad public communication about this change on social media or community channels? Many bots, integrations, or custom frontends may still be relying on the legacy bridge, and awareness is key to avoiding user errors or fund loss

1 Like
3

Hi team, thanks for putting together this important proposal and for clearly outlining the motivation and technical plan to disable the legacy USDT bridge.

As a regular user and community participant, I just want to kindly suggest that some additional effort be made on the communications side to help users transition smoothly. While the technical rationale is sound, many users may still be unaware of the new USDT0 standard or unsure how to swap from legacy USDT to USDT0 or how to exit to L1 if needed.

Would it be possible to:

  • Share recommended routes or tools for converting legacy USDT to USDT0 (e.g., official or community-trusted bridges or swap interfaces)?
  • Provide guidance for smart contract integrators who may still interact with the legacy bridge?
  • Include these in a public-facing FAQ or announcement campaign prior to final deprecation?

Good communication will go a long way in reducing friction and preventing user frustration or fund loss. This is the only area I feel could be improved in an otherwise well-considered proposal. Thanks again for the work you’re doing to make Arbitrum more robust and secure.

2 Likes
4

I am all in for completing the transition to the new USDT0 standard as the old bridge’s long wait time was a big obstacle for many and the biggest reason why people rely on third-party bridges, and the proposal does fully outline how this will be executed on Arbitrum.

However, would it be possible to add an estimated timelines of events for how this would be handled communications wise? As both @cp0x @cypherbadger mentioned, a lot of stakeholders, liquidity providers, etc. still use it, so while transactions will revert it would be best to give them proper notice.

I am merely suggesting making a announcement with the exact time and date the legacy bridge will be disabled that can be shared beforehand via Arbitrum’s channels, social media, perhaps even reaching out to Tether, and having a few media outlets report on it so users are aware of the change.

2 Likes
5

Yes, fully agree deploying this should me accompanied by - at least - an FAQ that can be widely shared ahead of time.

1 Like
6

Hey ! Just wanted to jump into the discussion because I believe there is a misunderstanding here. The USDT0 bridge mentioned is a solution developed by Tether; it does not use Arbitrum’s canonical (official) bridge. There is no need to submit this implementation to a vote, as it is not a product controlled by the Arbitrum DAO.

With a quick search on X, I was able to find official announcements from the Arbitrum account:

https://x.com/arbitrum/status/1884270274008453607

image
image1222×1193 128 KB

image
image1196×703 71.8 KB

Additionally, in the same Mirror article that was shared in this post, the following is stated:

image
image1515×1198 134 KB

image
image1471×244 21.1 KB

I hope my answer is helpful

3 Likes
7

Great clarification, thanks! I think what confused me was the part about users potentially losing funds. Technically, if the legacy USDT contract has already been upgraded, users shouldn’t be able to transfer or use it anymore, which is why the warning initially alarmed me a bit.

That said, it’s now clear that the main action left is to safely deactivate the Arbitrum USDT legacy bridge for users.

Still, I’d appreciate a bit more clarity:
Are users currently at any risk if they use the official Arbitrum USDT bridge?
I’m not even sure whether it’s still usable after the upgrade or not.

8

From a legal and operational risk standpoint, keeping the legacy USDT bridge live introduces unnecessary liability and exposure. The current configuration not only delays user access but risks non-compliant handling of smart contract deposits, especially where refund mechanisms are absent. Disabling the bridge streamlines compliance, protects user funds, and aligns with industry best practices in consumer asset safety and protocol governance. It’s a preventive legal upgrade as much as a technical one. Fully support this move.

1 Like
9

Blockworks Advisory supports this proposal at the Snapshot level.

As others have mentioned, keeping the remaining legacy USDT bridge up seems to be a liability in the long run with little to nothing to gain for the network nor the DAO. That coupled with its involvement between the Arbitrum canonical bridge make it a problem at Arbitrum’s front door from a logistics standpoint. We’re in favor of disabling it altogether.

10

Support this proposal.

The migration of USDT has moved well and has been a success. This step helps remove confusion and potential pain for individual users through unintended usage of the old bridge.

11

We support this proposal.
With Tether’s shift to the USDT0 implementation, there’s little to justify maintaining the legacy bridge.

12

The following reflects the views of L2BEAT’s governance team, composed of @krst, @Sinkas, and @Manugotsuka, and it’s based on their combined research, fact-checking, and ideation.

We reviewed the proposal with the help of L2BEAT’s research team, and we wanted to share our internal discussion to potentially help delegates better assess the proposal. Currently, USDT on Arbitrum has been upgraded to USDT0, so all USDT value on Arbitrum, as far as L2BEAT is concerned, is considered externally bridged, as USDT0 is a LayerZero OFT token (see risks associated with OFT tokens here).

If the proposed upgrade passes, it will prevent further bridging of Ethereum USDT to Arbitrum through the canonical custom gateway. Currently, most deposits flow through the LayerZero external bridge anyway, although there are indeed some sporadic deposits through the canonical bridge even after the migration (e.g., L1 deposit and L2 mint).

There are currently about $140,000 USDT in the gateway, while there’s more than $1,000,000,000 in the LayerZero lockbox. As you can see from the L2 mint transaction, canonically deposited USDT tokens are already minted as USDT0, so the Arbitrum holder trust assumption remains unchanged.

Disabling the legacy bridge will prevent users from using it, avoid the seven-day delay, and prevent any funds from being lost if the bridge is used by a smart contract on behalf of a user. These changes result in a better user experience with minimal effort.

Pushing users to an external bridge, however, is not entirely harmless, and it’s essential to unpack the nuance so delegates can make an informed decision. For example, if all USDT is in the L1 LayerZero lockbox, then withdrawing USDT depends on the DVN and the liveness of the lockbox owner. That trust assumption does not exist for canonically bridged tokens, making it less risky.

Additionally, to further clarify what @MinistroDolar mentioned above, this proposal aims to disable the legacy bridge and remove it from the gateway controller, which the Arbitrum DAO controls. It does not seek to replace the bridge in the gateway controller with the USDT0 bridge by Tether. Essentially, users will be ‘pushed’ to bridge via the external USDT0 bridge by Layer Zero.

As far as we’re concerned, the outcome of this proposal is largely irrelevant, as its passage would essentially be a rubber stamp of a decision that Tether has already made. Arbitrum and the Arbitrum DAO had no involvement in the decision to move to USDT0 — the DAO’s only control is over the gateway controller. In our opinion, it doesn’t make sense to maintain the legacy bridge when 99% of the liquidity is on the Layer Zero bridge.

However, while we’re not opposed to disabling the legacy bridge (as it is sensible from a practical standpoint), we’re also not comfortable endorsing the USDT0 setup and the liveness and safety trust assumptions that come with it.

As a result, we’ve decided that, if and when the proposal goes to a vote, we’ll abstain.

3 Likes
13

AranaDigital supports this proposal because the legacy gateway’s seven-day exit delay is no longer acceptable when USDT0 delivers near-instant transfers under the same contract address. Liquidity is already gone: only about $140k remains in the canonical bridge, while roughly a billion dollars now sits in the LayerZero lockbox, so the DAO is bearing risk without utility. Disabling the gateway also removes the auto-withdraw behaviour that can strand funds when smart contracts deposit for users. That said, we echo @cp0x, @cypherbadger, and the L2BEAT team (@krst, @Sinkas, @Manugotsuka) in requesting a clear public timeline, an FAQ, and explicit instructions for converting legacy USDT before execution, plus confirmation that any residual balance will be returned safely. Finally, because USDT0 relies on up-gradable LayerZero verifiers, we ask the proposers to outline the current security stack and the governance process for changing it so delegates can monitor ongoing risk.