Ethereum researchers are now using artificial intelligence agents to analyze the unexpected effects of Glamsterdam on existing contracts. This initiative comes after the postponement of the planned test on Hoodi, but it does not mean that the problem is already resolved.
In brief
- Ethereum postpones Glamsterdam test on Hoodi after technical risks discovered.
- ChainSecurity identifies at least 13 contracts that may be affected by the update.
- AI agents support researchers in detecting other unforeseen consequences.
- The teams coordinate their analyzes in order to strengthen the security of existing contracts.
- Ethereum extends this approach to Hegota to anticipate the risks of its future updates.
Thirteen exposed contracts lead to the postponement of the test
Glamsterdam’s dress rehearsal for Hoodi was due to take place on October 26. Developers ultimately postponed it after discovering risks associated with proposals EIP-8037 and EIP-8038, which change the way Ethereum charges for certain operations related to the creation of data on the blockchain.
The Swiss company ChainSecurity has identified at least 13 contracts already deployed whose operation could be disrupted. The problem appears when a contract calls another program and runs out of gas, the unit used to measure and bill calculations on Ethereum. In some cases, a normally canceled operation could leave permanent effects or prevent a subsequent attempt from succeeding.
The main known elements show that the deployment remains suspended:
- Glamsterdam test on Hoodi, scheduled for October 26, postponed;
- The risks concern the EIP-8037 and EIP-8038 proposals;
- ChainSecurity has spotted at least 13 potentially affected contracts;
- No financial losses linked to these flaws have been reported;
- Ethereum has not announced a new date for Hoodi or a final schedule for the mainnet.
Glamsterdam had already been activated on Sepolia on October 6. Hoodi was to constitute the second and final phase of testing before a possible arrival on the main network. The postponement therefore does not represent the abandonment of the update, but a pause intended to measure its consequences on existing applications.
Ethereum users do not need to take any urgent actions. The disputed changes are not active on the mainnet and the vulnerabilities were discovered before final deployment.
AI agents coordinate contract analysis
Researchers from Eth Labs and the Ethereum Foundation separately used AI agents to examine contracts that might be affected. After discovering that they were following a similar approach, they created a GitHub repository to serve as a communications board for their automated systems.
These agents can go through large amounts of code, simulate different behaviors, and report contracts whose operation depends on old gas rules. Their mission relates in particular to the “second order effects”that is to say indirect consequences which do not appear immediately in the technical proposal.
Derek Chiang, a member of Eth Labs, compared this method to that used by malicious agents in a Hugging Face incident. He said of the coordination table:
It turns out that the same model is also useful for good agents.
This formula should not, however, lead to AI being presented as the author of the rescue. ChainSecurity detected the risks that caused the postponement. Agents now step in to expand the analysis, search for other relevant contracts, and help human specialists examine more scenarios.
Chiang recognizes also that the problem was identified very late, which forced the teams to react urgently. According to him, protocol designers should analyze much earlier the dysfunctions that their updates could introduce.
Ethereum tests new continuous auditing model
The experience could change the way Ethereum prepares for its next developments. Traditional audits rely on human teams, automated tools and testing periods. AI agents add constant exploration capability, with multiple systems able to share their results and distribute tasks.
The device has already been extended to the EIP-8198, nicknamed “Quick Slots”. This proposal, considered for the next update called Hegota, seeks to shorten the time between blocks. A second coordination platform was opened in order to examine the consequences before its development reaches an advanced phase.
This early use could avoid reproducing the situation in Glamsterdam. However, it has its own limitations. An agent can produce false signals, misinterpret how a contract works, or overlook an important interaction. Each alert must therefore be reproduced, understood and validated by human researchers.
A study published in 2026 on the security of Ethereum contracts has watch that advanced agents could detect, correct, and also exploit end-to-end vulnerabilities. This dual capacity reinforces their interest in audits while imposing controlled environments and strict monitoring.
The next Glamsterdam test will now depend on the conclusions of this analysis. So the real progress lies not yet in a patched update, but in the emergence of a model where humans and AI agents examine risks together before they reach the mainnet.
Maximize your Tremplin.io experience with our ‘Read to Earn’ program! For every article you read, earn points and access exclusive rewards. Sign up now and start earning benefits.
