温馨提示：本站仅提供公开网络链接索引服务，不存储、不篡改任何第三方内容，所有内容版权归原作者所有
AI智能索引来源：http://www.bee.com/zh_cn/61922.html
点击访问原文链接

AI has successfully simulated the theft of $4.6 million and has learned to autonomously attack smart contracts. | Bee Network

进入原网站 导航首页
AI has successfully simulated the theft of $4.6 million and has learned to autonomously attack smart contracts. | Bee Network Login 热门新闻 备忘录启动板 人工智能代理 德西 TopChainExplorer 给 Newbee 100 倍金币 蜜蜂游戏 重要网站 必备应用程序 加密货币名人 德平 新手必备 陷阱探测器 基本工具 高级网站 交流 NFT 工具 你好、, 签出 Web3 宇宙 游戏 DApp 蜂巢 成长平台 生态 搜索 英语 充值金币 登录 下载 Web3 大学 游戏 DApp 蜂巢 生态 家•分析•正文 AI has successfully simulated the theft of $4.6 million and has learned to autonomously attack smart contracts.分析3 个月前发布怀亚特 15,647 16 Author｜Azuma ( @azuma_eth )

Anthropic, a leading AI company and developer of the Claude LLM model, today announced a test that uses AI to autonomously attack smart contracts (Note: Anthropic was invested in by FTX, and theoretically its equity value is now enough to cover the FTX asset vulnerabilities, but it was sold off at a low price by the bankruptcy administration team).

The final test results show that profitable and reusable AI autonomous attacks are technically feasible. It’s important to note that Anthropic’s experiments were conducted only in a simulated blockchain environment and were not tested on a real blockchain, therefore they did not affect any real-world assets.

Below, we will briefly introduce Anthropic’s testing scheme.

Anthropic first built a smart contract exploitation benchmark (SCONE-bench), the first benchmark in history to measure the exploitation capabilities of AI agents by simulating the total value of stolen funds. That is, the benchmark does not rely on vulnerability bounties or speculative models, but directly quantifies the loss and assesses the capability through changes in on-chain assets.

SCONE-bench uses 405 real contracts that were attacked between 2020 and 2025 as a test set, located on three EVM chains: Ethereum, BSC, and Base. For each target contract, an AI Agent running in a sandbox environment attempts to attack the specified contract within a limited time (60 minutes) using tools exposed by the Model Context Protocol (MCP). To ensure the reproducibility of results, Anthropic built an evaluation framework that uses Docker containers for sandboxing and scalable execution. Each container runs a local blockchain forked at a specific block height.

The following are Anthropic’s test results for different scenarios.

First, Anthropic evaluated the performance of 10 models—Llama 3, GPT-4o, DeepSeek V3, Sonnet 3.7, o3, Opus 4, Opus 4.1, GPT-5, Sonnet 4.5, and Opus 4.5—on all 405 benchmark vulnerable contracts. Overall, these models generated ready-to-use exploit scripts for 207 of them (51.11%), simulating the theft of $550.1 million. Secondly, to control for potential data contamination, Anthropic evaluated 34 contracts attacked after March 1, 2025, using the same 10 models— this date was chosen because March 1st is the knowledge expiration date for these models . Overall, Opus 4.5, Sonnet 4.5, and GPT-5 successfully exploited 19 of them (55.8%), simulating a maximum theft of $4.6 million; the best-performing model, Opus 4.5, successfully exploited 17 of them (50%), simulating a theft of $4.5 million. Finally, to evaluate the AI Agent’s ability to discover new zero-day vulnerabilities, Anthropic had Sonnet 4.5 and GPT-5 evaluate 2,849 recently deployed contracts with no known vulnerabilities on October 3, 2025. Each AI Agent discovered two new zero-day vulnerabilities and generated attack schemes worth $3,694, with GPT-5’s API costing $3,476. This demonstrates that profitable, real-world reusable AI-driven attacks are technically feasible. After Anthropic released its test results, many well-known figures in the industry, including Haseeb, managing partner of Dragonfly, marveled at the astonishing speed at which AI has progressed from theory to practical application.

But just how fast is this speed? Anthropic has provided the answer.

In its test conclusion, Anthropic stated that in just one year, the percentage of vulnerabilities that AI could exploit in this benchmark test skyrocketed from 2% to 55.88%, and the amount of money that could be stolen surged from $5,000 to $4.6 million. Anthropic also found that the value of potential exploitable vulnerabilities roughly doubles every 1.3 months, while the cost of tokens decreases by about 23% every two months—in the experiment, the average cost of having an AI agent perform an exhaustive vulnerability scan of a smart contract is currently only $1.22.

Anthropic states that in 2025, over half of all real attacks on the blockchain—presumably carried out by skilled human attackers—could have been accomplished entirely autonomously by existing AI agents. As costs decrease and capabilities compound, the window of opportunity before vulnerable contracts are exploited after deployment on the chain will continue to shrink, leaving developers with less and less time for vulnerability detection and patching… AI can be used to exploit vulnerabilities, but it can also be used to patch them . Security professionals need to update their understanding; it’s time to leverage AI for defense.

本文来源于互联网： AI has successfully simulated the theft of $4.6 million and has learned to autonomously attack smart contracts.

 

Related: Is an on-chain “subprime crisis” already emerging? The path to maturity for DeFi structured products.

Original translation by: AididiaoJP, Foresight News The Rise of Risk Management and On-Chain Capital Allocator (OCCA) DeFi has entered a new structured phase, with institutional trading strategies being abstracted into composable and tokenizable assets. It all began with the emergence of liquidity-staking tokens, and Ethena Labs’ tokenized basis trading became a key turning point for DeFi structured products. This protocol packaged a delta-neutral hedging strategy, which required 24-hour margin management, into a synthetic dollar token, allowing users to participate with a single click, thus redefining their expectations of DeFi. What was once a product reserved for trading firms and institutions has now entered the mainstream. USDe has become the fastest stablecoin to reach a total value locked of $10 billion. Ethena’s success confirms the strong market demand for “institutional strategy…

 

#分析# 以太坊# Sandbox# 代币# 工具© 版权声明文章版权归作者所有，未经允许请勿转载。 上一篇 Kalshi teams up with Solana: Is the US’s first compliant prediction market starting to reap profits from crypto enthusiasts? 下一篇 Every country is heavily indebted, so who are the creditors? 相关文章 Nietzsche Penguin Surges 6000x in a Week, Is Solana Meme Coin Reclaiming Its Glory? 6086cf14eb90bc67ca4fc62b 9,051 1 In-depth analysis of the truth behind xUSD’s de-pegging: The domino crisis triggered by the October 11 crash. 6086cf14eb90bc67ca4fc62b 1,161,184 4659 The Rise and Future of Perp DEX: A Structural Revolution in On-Chain Derivatives 6086cf14eb90bc67ca4fc62b 11,621 Trump’s Shopping Cart: Greenland, Canals, and Canada 6086cf14eb90bc67ca4fc62b 10,967 2 The truth of the data: 2/3 of the Meme coins promoted by KOLs have returned to zero, and only 1% of the tokens have incr 6086cf14eb90bc67ca4fc62b 38,905 2 Six charts analyze ETH chip distribution as new highs approachRecommended Articles 6086cf14eb90bc67ca4fc62b 25,396 4 Bee.com 全球最大的 Web3 门户网站 合作伙伴 硬币卡 Binance CoinMarketCap CoinGecko Coinlive 装甲 下载蜜蜂网络APP，开始web3之旅 白皮书 角色 常见问题 © 2021-2026.保留所有权利。. 隐私政策 | 服务条款 下载蜜蜂网络 APP 并开始 web3 之旅 全球最大的 Web3 门户网站 合作伙伴 CoinCarp Binance CoinMarketCap CoinGecko Coinlive Armors 白皮书 角色 常见问题 © 2021-2026.保留所有权利。. 隐私政策 | 服务条款 搜索 搜索InSite链上社会新闻 热门推荐： 空投猎人 数据分析 加密货币名人 陷阱探测器 简体中文 English 繁體中文 日本語 Tiếng Việt العربية 한국어 Bahasa Indonesia हिन्दी اردو Русский 简体中文

智能索引记录