Daily · AI Infrastructure and Cybersecurity Crisis · November 2, 2026

Key points

Binance Reopens After Record Hack

Binance resumed withdrawals on November 1 following a US$6.3 billion theft attributed by the FBI to North Korea's Lazarus Group, also known as TraderTraitor. The exchange processed $4.65 billion in withdrawals and $1.01 billion in deposits between 6:00 a.m. and 3:30 p.m. UTC, resulting in net outflows of $3.64 billion, or 2.1% of its total assets.

The company stated it has covered 95% of the stolen assets using its corporate treasury and reserve funds, up from 90% in previous updates. Preliminary findings suggest the intrusion began with a compromised engineer's laptop. Binance CEO Richard Teng confirmed the exchange will cover all customer losses, while founder Changpeng Zhao noted that 98% of users chose to stay despite the initial panic.

The incident marks the largest known cryptocurrency theft to date. Analysts at TRM Labs reported that attackers stole US$2.9 billion across nearly 150 incidents last year, highlighting a persistent threat vector for digital asset custodians.

Anthropic IPO and Compute Commitments

Anthropic filed an amended Form S-1 with the SEC, reporting $33.24 billion in revenue for the nine months ended September 30, a 1,200% increase year-over-year. The company reported a net loss of $104 billion, largely driven by fair value changes in convertible notes, and disclosed aggregate purchase commitments of approximately $895.4 billion for compute and infrastructure.

The filing highlights the massive capital intensity of frontier AI development, with compute expenses rising 329% over the same period. Anthropic is targeting a $2 trillion valuation and plans to raise up to $3.5 billion from Japanese investors. The prospectus warns that if revenue growth does not outpace these commitments, the company could face default risks.

This filing sets a new benchmark for AI sector transparency, revealing the scale of infrastructure spending required to maintain competitive advantage in model training and inference.

European Retail Cyberattack

A cyberattack on November 2 disrupted refrigeration systems in supermarkets across Sweden, Belgium, Luxembourg, Ireland, and Estonia. In Stockholm, an ICA store saw freezers reach 50 degrees Celsius, while in Belgium, approximately 40 Delhaize stores and 15 Carrefour stores were affected. Stephan Moelans, CEO of Belgian refrigeration firm Sabcobel, confirmed the attack involved switching evaporators to defrost mode.

The incident follows a similar attack in France in October that affected hundreds of supermarkets. Sweden's national cybersecurity centre identified a large number of exposed operational technology devices requiring updates. The attacks have resulted in significant food waste and financial losses, with one Belgian butchery reporting €15,000 in ruined meat.

AI Infrastructure Financing

Oracle filed plans to issue $10 billion in common stock and $10 billion in senior notes to finance its AI cloud buildout. The company's shares fell more than 4% on the announcement, extending a 33% decline in 2026. Oracle's five-year credit default swap spread has risen 270% in one year, approaching high-yield levels.

Simultaneously, Marvell Technology announced an all-stock acquisition of Lumentum for an implied enterprise value of $115 billion, citing a fundamental shift toward optical connectivity in AI infrastructure. The deal aims to consolidate electrical and optical connectivity under one vendor as data center bandwidth demands increase.

AI Agent Adoption and Security

JPMorgan Chase, Amazon, and other major retailers are blocking personal AI agents like Meta Muse from their platforms, citing security risks. An audit of 51 U.S. retailers found that only 11 had taken a deliberate position on AI agents, with five explicitly restricting them. Conversely, Walmart announced it is opening its platform to customer-preferred agents.

Security researchers at Wiz identified a wave of October cyberattacks on at least nine AI development platforms, dubbed "Black October." Attackers targeted AI access credentials and in some cases deployed autonomous AI agents. Notable incidents include a breach of OpenCode affecting 1.4% of users and a malicious LiteLLM release that stole credentials from PyPI.

Regional AI and Energy Policies

South Korea proposed a record 2027 budget of 820.9 trillion won, allocating 21.3 trillion won to semiconductors and AI. The budget includes a 4.7 trillion won fund for a sovereign AI model. In the UAE, OpenAI signed four agreements with Abu Dhabi entities to deploy AI in energy, government, and investment sectors, co-leading a $20 billion funding round.

Germany's government AI task force drafted a report calling for the country to become a "sovereign AI nation" by 2030. Meanwhile, China's CCID reported that AI inference tokens have surpassed training tokens for the first time, with daily token calls exceeding 175 trillion.

Our read

The convergence of massive capital commitments in AI infrastructure and escalating cyber threats against digital assets creates a dual risk landscape for decision-makers. Anthropic's $895 billion compute commitment and Oracle's $20 billion raise signal that the industry is leveraging debt and equity at unprecedented scales, raising questions about long-term sustainability and credit risk. Simultaneously, the Binance hack and the "Black October" attacks on AI platforms demonstrate that security vulnerabilities in both financial custody and developer tooling are becoming primary vectors for systemic loss. Organizations must now integrate rigorous counterparty risk assessment with advanced operational technology security to protect both capital and data integrity.

This material was produced automatically by a large-language-model system from the public sources listed below; it is AI-generated content and may contain inaccuracies — verify facts against the original sources.

Sources