News linked to both this project and an event.
According to Cointelegraph, Ethereum researcher Justin Drake warned that the rapid advancement of AI could accelerate mathematical attacks against the Elliptic Curve Digital Signature Algorithm (ECDSA), potentially breaking existing cryptographic systems within months under the worst-case scenario, and called on the industry to initiate "bunker mode"—transferring assets to new addresses with previously unexposed public keys to mitigate risk. Several industry insiders, however, expressed reservations. Coinbase Chief Cryptographer Yehuda Lindell stated there is currently "no evidence" that the mathematical assumptions underpinning ECDSA have been weakened; Trezor Bitcoin analyst Lucien Bourdon similarly noted that migrating assets could backfire and inadvertently create opportunities for scams. MetaMask founder Taylor Monahan was even more direct, asserting that if elliptic curve cryptography were truly compromised, "the entire world would go to zero," rendering bunker mode ineffective. Meanwhile, Ethereum has laid out a post-quantum roadmap based on hash-based signatures and ZK proofs, while public blockchains such as Solana and NEAR are also exploring their own quantum-safe alternatives, though concrete deployment timelines remain unset.
Odaily News: Dragonfly Managing Partner Haseeb Qureshi stated that the "bunker mode" proposed by Ethereum researcher Justin Drake to address the risk of AI cracking cryptographic signatures amounts to "cryptographic doomerism," and argued that migrating tokens to new addresses is not a real solution.Qureshi said that "bunker mode" can only protect assets that users have not yet transferred out of new addresses. If tokens in other addresses are attacked and heavily sold off, those assets could still lose value. He suggested that blockchain networks adopt a "Cryptographic Recovery Mode," using hash-based backup signature schemes that allow users to bind recovery mechanisms in advance, so that validators can assist in recovering assets after cryptographic signatures are cracked.Previously, Ethereum researcher Justin Drake warned that with the rapid improvement of AI mathematical capabilities, the ECDSA signature algorithm protecting crypto wallets could face the risk of being cracked within "months rather than years," and advised users to gradually migrate funds to new wallets that have not exposed public keys. Glassnode data shows that approximately 6.26 million BTC are currently held in addresses with potential risk. (Cointelegraph)
Odaily News: As the crypto industry discusses whether users need to enter "bunker mode" before AI or quantum computing technologies expose wallet private keys, custody institutions face a more complex challenge: how to upgrade key management, approval, and audit systems before an emergency migration occurs.Project Eleven CEO Alex Pruden stated that institutions have already begun preparing for post-quantum migration outside of blockchain, and different blockchains may adopt different quantum-resistant signature schemes and migration paths in the future. Institutions holding multiple types of assets will face the challenge of simultaneously adapting to multiple cryptographic systems.Quantus co-founder and CEO Christopher Smith stated that AI is accelerating the development of quantum hardware and software, and the tail risk of quantum attacks should be incorporated into portfolio decisions — this is an institutional fiduciary duty.Currently, the timeline for quantum computers to crack Bitcoin and Ethereum's existing public-key cryptographic systems remains uncertain, but the industry is gradually viewing the coming years as a window for establishing migration plans, rather than waiting for an attack to occur before responding reactively. (CoinDesk)
Odaily News: According to blockchain security firm Blockaid's monitoring, More Markets (More Labs) on Flow EVM has been exploited. The attacker leveraged Ankr's liquid staking tokens and the E-Mode mechanism to drain More Markets' WFLOW lending reserves, transferring approximately 15.5 million WFLOW from the mFlowWFLOW reserve. The detected impact amount is approximately $9.3 million, and the related attack transaction cluster also includes post-exploit fund transfer operations. At present, the specific losses and the destination of the attacker's funds are still under further confirmation.
Coldcard has released the latest firmware 5.6.1 (Mk4/Mk5) and 1.5.1Q (Q). This update is based on a three-week security review following the emergency fix on July 31, addressing security risks brought by the previous mnemonic generation attack. Each newly generated mnemonic must now include at least one source of user entropy, including at least 65 irregular keystrokes, 50 physical dice throws, or 128 physical coin flips, combined with fresh entropy provided by STM32 TRNG, SE1, and SE2. The new firmware also adds instant staged PSBT verification before signing, strengthens USB connection and firmware update boundaries, improves Delta Mode isolation mechanisms, fixes active wallet backup issues, enhances random number generator initialization and fault checking, adjusts SIGHASH default settings, and includes multiple security and correctness improvements. Coldcard stated that this update aims to further reduce the risk of device attacks. The official reminder notes that updating the firmware cannot fix existing mnemonics generated by previously affected firmware. If users' mnemonics fall within the scope of this security advisory, they should first update the device, then generate and verify a completely new mnemonic, and migrate funds to the new wallet. Coldcard recommends that all Mk4, Mk5, and Q users update their devices promptly and verify the signature of the downloaded firmware.
Anthropic 宣布其 Opus 5 模型在浏览器智能体场景中几乎免疫提示词注入攻击。在 129 个测试场景中，攻击成功率为零；在 Gray Swan 通用提示词注入测试中，15 次尝试后的成功率从 Opus 4.8 的 5.5% 降至 2.0%。零成功率仅在 Claude Cowork 等产品开启 Auto Mode 时实现，该模式叠加了输入扫描与执行拦截两层防御。提示词注入被视为 AI 智能体面临的最大安全隐患之一，此次改进或标志着该问题在特定场景下得到有效缓解。