if you could delete ONE platform forever
the one that leaves most of their users broke
which one are you picking?
the one that leaves most of their users broke
which one are you picking?
10
1
14
3,656
ICYMI 🚨
BlackRock ETF wallets transferred over $355Million in bitcoin:native and ethereum:native to Coinbase Prime within the past 2 hours.
– 3,215 bitcoin:native $262.41M
– 38,882 ethereum:native $93.22M
BlackRock ETF wallets transferred over $355Million in bitcoin:native and ethereum:native to Coinbase Prime within the past 2 hours.
– 3,215 bitcoin:native $262.41M
– 38,882 ethereum:native $93.22M
5
2
15
2,790
Bitcoin Core, the software most of the bitcoin network runs on, used to keep medium and high severity flaws under embargo until two weeks after the last affected version reached end of life, which could mean about a year.
Maintainers have now merged a change to the project's ds this for bugs an AI can easily rediscover.
During triage, Core can now flag a vulnerability as easy to discover, meaning one that can be found with a simple prompt to an LLM.
Those fixes get backported to every supported version and shipped in a point release as soon as possible, together with a security advisory.
Antoine Poinsot of Chaincode Labs, who wrote the change, pointed to a jump in reports driven by LLMs lowering the bar to find certain classes of issues.
In late July, a 16-developer volunteer group ran AI models across 390 bitcoin projects and logged 4,962 potential flaws, 85 of them critical.
The tradeoff is less testing:
patches won't get as much "bake time" before reaching users.
No bug has been flagged under the new rule yet.
Maintainers have now merged a change to the project's ds this for bugs an AI can easily rediscover.
During triage, Core can now flag a vulnerability as easy to discover, meaning one that can be found with a simple prompt to an LLM.
Those fixes get backported to every supported version and shipped in a point release as soon as possible, together with a security advisory.
Antoine Poinsot of Chaincode Labs, who wrote the change, pointed to a jump in reports driven by LLMs lowering the bar to find certain classes of issues.
In late July, a 16-developer volunteer group ran AI models across 390 bitcoin projects and logged 4,962 potential flaws, 85 of them critical.
The tradeoff is less testing:
patches won't get as much "bake time" before reaching users.
No bug has been flagged under the new rule yet.
4
4
16
1,883
Ethereum Postponed Glamsterdam on Hoodi Testnet, No new date set.
Core developers agreed on ACDE call #248 to pull the Glamsterdam activation on the Hoodi testnet, which had been scheduled for October 26 at epoch 132,352.
The goal is to further investigate how EIP-8037 gas xisting mainnet smart contracts.
The rollback removes both the Amsterdam execution layer and Gloas consensus layer activations, plus a gas-limit schedule that would have lifted Hoodi's limit to 200M.
Hoodi’s chain history is not reverted, and the updated configuration contains no replacement epoch or timestamp.
Hoodi was the next test step before Glamsterdam reaches mainnet, and no mainnet date has been set.
Node and validator operators need to update execution and consensus clients before October 26. Besu 26.10.1, Nethermind 2.2.1 and Lodestar v1.50.1 already ship the revised config, while other clients will follow.
No action is required from mainnet users or ETH holders.
Core developers agreed on ACDE call #248 to pull the Glamsterdam activation on the Hoodi testnet, which had been scheduled for October 26 at epoch 132,352.
The goal is to further investigate how EIP-8037 gas xisting mainnet smart contracts.
The rollback removes both the Amsterdam execution layer and Gloas consensus layer activations, plus a gas-limit schedule that would have lifted Hoodi's limit to 200M.
Hoodi’s chain history is not reverted, and the updated configuration contains no replacement epoch or timestamp.
Hoodi was the next test step before Glamsterdam reaches mainnet, and no mainnet date has been set.
Node and validator operators need to update execution and consensus clients before October 26. Besu 26.10.1, Nethermind 2.2.1 and Lodestar v1.50.1 already ship the revised config, while other clients will follow.
No action is required from mainnet users or ETH holders.
7
3
18
1,370
memecoins used to prioritize community growth
now the dominant strategy is just launching as many as possible
what’s the fix?
now the dominant strategy is just launching as many as possible
what’s the fix?
7
2
18
1,833
Unbacked inflation has hit two of Swell DAO’s liquid staking tokens
the admin keys for swETH and rswETH appear to be compromised, with an attacker pushing the tokens’ NAV higher by up to 1% every hour.
The keys control proxy admin permissions, which lets the holder point the s at new implementation contracts.
Four malicious upgrade contracts have already been deployed, although none has been queued through the multisig yet.
If executed, the malicious upgrades could drain roughly 2,500 ETH worth over $6million.
The root cause is unconfirmed.
A possible link to the cloud deployment platform Railway was floated based on timing alone, with no conclusive evidence.
Holders of swETH and rswETH face the risk of trading against an artificially inflated exchange rate until the keys are secured.
the admin keys for swETH and rswETH appear to be compromised, with an attacker pushing the tokens’ NAV higher by up to 1% every hour.
The keys control proxy admin permissions, which lets the holder point the s at new implementation contracts.
Four malicious upgrade contracts have already been deployed, although none has been queued through the multisig yet.
If executed, the malicious upgrades could drain roughly 2,500 ETH worth over $6million.
The root cause is unconfirmed.
A possible link to the cloud deployment platform Railway was floated based on timing alone, with no conclusive evidence.
Holders of swETH and rswETH face the risk of trading against an artificially inflated exchange rate until the keys are secured.
3
2
18
784
