A research report co-authored by Accenture and IE University on artificial intelligence in African agriculture was found to contain false citations, misattributed research and possibly nonexistent case studies. According to Sina Finance, AI detection company GPTZero identified factual errors and 15 citation problems, including inaccurate paper titles and author names. The report, published in November 2025, cited two seemingly nonexistent organizations under different names and claimed predictive AI detected cacao swollen shoot disease with 99% accuracy in Colombia and Ecuador. Its supporting reference focused on West Africa and did not mention either South American country. Other errors included incorrect rankings for Ghana and Uganda in the University of Oxford’s 2023 AI Readiness Index. Accenture said its review found the errors resulted from human mistakes and that there was no evidence they were caused by AI hallucinations. The company said it would correct and republish the report, but declined to explain the basis for its conclusion or how the errors occurred. IE University said the authors were reviewing the sources and data and would make necessary corrections if errors were confirmed. GPTZero co-founder Edward Tian said the errors largely matched patterns typical of AI hallucinations. GPTZero said it manually checked each error it flagged and conducted additional verification using open-source intelligence. The company’s AI detection tools have also faced criticism for sometimes incorrectly labeling human-written content as AI-generated.

Daren Matsuoka, an investment partner at a16z Frontier, argues that AI's ability to reproduce software shifts Web3's opportunity from applications to reusable networks. According to Foresight News, Matsuoka says AI has driven software development costs close to zero, making it easier for developers to build prototypes and connect them to existing protocols. He says this change revives the potential of crypto concepts such as NFTs, DAOs, decentralized social networks and on-chain games. Their value, he argues, depends on people and AI agents building on shared networks rather than on software alone. Uniswap illustrates composability by allowing new applications to use existing liquidity instead of creating their own from scratch. Matsuoka argues that AI can reproduce application interfaces and APIs, but not the underlying networks' accumulated liquidity, identity, reputation, social connections, history and trust. As software becomes easier to recreate, he says, access to these shared resources becomes more important. A person can generate a wallet or an NFT with AI, but cannot thereby create deep liquidity, an NFT's provenance or its community. He describes blockchains as potential shared state layers for the internet, where ownership, identity, liquidity, reputation and social relationships remain scarce resources. In his view, building reusable networks is a major opportunity for Web3 developers.

Bitcoin’s Seller Exhaustion Model has fallen to its lowest level since the bear market. According to ChainCatcher, CryptoQuant analyst Darkfost said the model compares the share of supply in profit with average monthly volatility to measure whether sellers have enough coins to keep pushing prices lower. BTC has fallen more than 10% from its previous high, and short-term profit-taking has largely run its course, weakening seller momentum. Historically, a sharp drop in selling pressure can help form a market bottom, Darkfost said, but easing pressure does not mean an immediate rebound. Without sufficient new demand and a meaningful return of buyers, Bitcoin’s upside remains limited.

Banks overseeing Turkey’s fund liquidations sold about $254 million of blue-chip shares held by frozen funds in off-market transactions, Bloomberg reported. The move could help remove an overhang on the country’s stock indexes.

Evercore ISI reiterated its Outperform rating on Apple and maintained its $380 price target after the company released its fiscal 2026 Form 10-K annual report on November 4. According to Sina Finance, the target implies about 14.5% upside from the latest share price of $331.90, while Apple’s manufacturing purchase commitments rose by approximately $33 billion from the previous quarter to $89.7 billion, with $82.7 billion due within the next 12 months. Evercore said the increase in commitments may be mainly related to purchases of memory chips, particularly NAND flash, reflecting Apple’s efforts to strengthen supply-chain reserves. Apple listed Siri AI for the first time as a service included with purchases of iPhone, Mac, and iPad devices, and deferred some device sales revenue for recognition during the period in which the related services are provided. Deferred revenue increased about 8% year over year to $14.8 billion, indicating that AI services have been incorporated into the company’s device sales and revenue recognition framework. Apple’s quarterly inventory rose by approximately $3.4 billion from the previous quarter to $14.5 billion. Component inventory increased by about $2.8 billion to $10.4 billion, up approximately 37% from the previous quarter, indicating that the company is building inventory for future hardware deliveries.

India’s Nifty 50 rose 0.17% to 23,850.35 on Wednesday, while Asian chip-heavy markets fell more than 3%. Tax-relief hopes, renewed foreign buying and the index’s lower exposure to AI hardware are supporting it, according to BeInCrypto. Global funds bought a net $2 billion of Indian stocks in the last two weeks of October, reversing the earlier rotation toward Korea and Taiwan. Brent crude near $103 a barrel and the lack of a tax-relief timeline remain risks.

According to Jin10, Industrial Bank Chief Economist Jia Yang Groen said Democratic control of the House could have the greatest impact on the U.S. Treasury market among the possible outcomes of the U.S. midterm elections. He said losing the House and its budgetary powers would make it difficult for the Trump administration to pursue further tax cuts or other expansionary fiscal policies. Groen said the resulting fiscal-policy stalemate would reduce the likelihood of further spending increases, lower the risk of continued fiscal deterioration and limit upward pressure on long-term Treasury yields. He added that the Democrats’ narrow congressional majority, the Senate filibuster rule and the president’s veto power would significantly constrain efforts to increase spending.

According to Jin10, MissionSquare Chief Investment Officer Shari Hensrud said market expectations for the October U.S. employment report were modest, but its key figures should be treated cautiously. Three of the past four reports were revised down, reducing payrolls by a cumulative roughly 150,000 jobs, while average monthly job gains over the past three months were about 50,000. Hensrud also noted that Brent crude briefly approached $118 a barrel in October and long-term Treasury yields fell below 5%. She said these factors might not be fully reflected in the report, and their effects on hiring were expected to emerge in November and December. She added that the negative impact of higher energy costs could outweigh the improvement from lower borrowing costs.

若民主党顺利赢得参众两院，该党可能发动一场“立法冲刺”，迅速通过一系列广受欢迎的法案，同时推翻特朗普最不受欢迎的政策。

在币安遭遇63亿美元黑客攻击后，韩国正考虑设立国家网络安全机构。该提案强调了日益严峻的网络威胁、人工智能驱动的攻击以及加强公私合作的必要性。

新加坡航空公司证实，其Amadeus数据泄露事件导致4.6万条预订记录被访问。个人旅行信息遭到泄露，引发了人们对航空公司供应链网络安全的担忧。

特朗普中期选举失利反映出美国选民对特朗普经济及内外政策不满，特朗普未来两年立法议程将面临更多制约。加密行业在选举中投入巨额资金，加密监管政策也将因此受更多关注。

近几年缺少天王级项目，不一定意味着行业衰退，反而可能是行业从野蛮生长走向成熟竞争的表现。

黑客用近五个月时间，完成了从工程师个人电脑到交易所核心钱包系统的渗透。

俄罗斯正准备开设首个有组织的加密货币交易市场，但国际制裁和对投资者需求的疑虑可能会使莫斯科交易所进军数字资产领域的最新举措变得复杂。

Binance,钱包,黑客攻击,论打工人能惹多大的祸？币安被盗竟因员工偷偷面试 金色财经,事到如今，最好奇的是币安会如何处理惹祸的打工人？

美国中期选举投票日，预测市场押注民主党横扫两院概率高达82%，行业合计押注规模超14.8亿美元创历史新高。此番预测结果是否应验，将直接决定这一新兴行业的公信力命运。

在Ubuntu和Red Hat系统中，一个近乎满分的Linux内核漏洞至今仍未修复。Anthropic未发布的一款大模型，成功地将这个漏洞，自动转化为可远程获取root权限的高危利用链！