AI Industry Development & Policy Dynamics

According to Reuters, on June 5, Foxconn raised its second-quarter guidance above previous expectations, citing strong demand for AI servers. The Taiwan-based company, Nvidia's biggest server maker and Apple's top iPhone assembler, said its AI server business rose more than 60 percent year on year i

According to Nvidia CEO Jensen Huang on June 5, the company's Taiwan supply chain will need to prepare for a scale significantly larger next half of 2027 than this year. Huang stated that Nvidia already partners with approximately 120 Taiwanese suppliers and has increased second-half 2026 supply

According to Beating, LM Studio officially launched Locally, its mobile application for iPhone and iPad. The app combines local on-device AI model execution with LM Link, a new remote access feature that allows users to connect their iPhones to desktop computers running larger models. The

According to Michael Saylor, CEO of MicroStrategy, Bitcoin's recent selloff may be driven by capital rotation toward artificial intelligence opportunities, rather than deteriorating fundamentals of Bitcoin

According to Beating, Google DeepMind released Magenta RealTime 2 (MRT2), an open-source real-time music generation model optimized for Apple Silicon MacBooks. The model enables musicians to control audio generation through MIDI signals, text prompts, or audio samples with latency below 200

According to Yonhapnews, Nvidia CEO Jensen Huang visited T1 PC bang (internet cafe) in Seoul on June 5 to meet professional esports player Faker (Lee Sang-hyeok) and gifted him the latest GeForce RTX 5090 GPU. During the visit, Huang joked about Faker's previous RTX 4070 graphics card, calling it "a

According to Beating, NVIDIA released two specialized models optimized for agent ecosystems today. Nemotron 3.5 ASR is a real-time multilingual speech recognition model with 600 million parameters supporting over 40 languages with streaming processing, delivering sub-100-millisecond latency. Nemotro

Arthur Hayes, co-founder of BitMEX exchange and crypto-focused Maelstrom Fund, dumped his entire Hyperliquid (HYPE) and Near protocol (NEAR) holdings ahead of the SpaceX IPO scheduled for next week. Hayes cited the upcoming IPOs of SpaceX, Anthropic, and OpenAI as the reason for rebalancing his

According to Bloomberg, Samsung, SK hynix, and Micron have been qualified to supply HBM4 memory for Nvidia's Vera Rubin AI platform, with all three currently in production. Vera Rubin is set for delivery in the third quarter. HBM4 doubles interface bandwidth to 2 TB/s compared with HBM3E's 1 TB/s, s

According to Chalootech, on June 5, embodied AI company Yuanli Lingji completed the acquisition of logistics robotics company Atomix through equity merger and secured new strategic funding from investors including Zhipu, Stanza, and

According to its latest technical blog post released on June 1, Anthropic disclosed that over 80% of its codebase is now generated by Claude AI, with engineer productivity increasing eightfold compared to 2024. The company warned that as AI systems develop recursive self-improvement capabilities,

According to Jin10 Data, Chinese Foreign Ministry spokesperson Mao Ning stated on June 5 that artificial intelligence should not become an exclusive domain of major powers or slide toward competition and confrontation. Mao noted that AI is profoundly transforming production and lifestyles and

According to The Information, Apple's revamped Siri, set to debut as soon as next week, may run partially on Nvidia Blackwell B200 GPUs provided by Google.

According to Seeking Alpha, billings revenue for AI-focused apps on Apple's App Store increased fourfold in 2025. Additionally, 40 of the top 100 apps on the storefront that year featured consumer-facing AI

According to Bloomberg, U.S. company Convalt Energy agreed to invest $6.2 billion to build a 1,200-megawatt hydropower project and an artificial intelligence data center in Lesotho on June 5. Lesotho government spokesperson Boitelo Rabele said in a statement that this is the country's largest

According to Cloudflare Radar, AI bots accounted for 57% of global HTML web traffic as of April 27, surpassing human users at 43%. Cloudflare CEO Matthew Prince stated the milestone occurred faster than anticipated; he had previously expected this shift in late 2027, then revised to early 2027,

According to OneMillion_AI, open-source database platform Supabase announced completion of a $500 million Series F funding round, valuing the company at $10.5 billion. The round was led by Singapore's GIC sovereign wealth fund, with participation from Stripe, Georgian, and Salesforce Ventures. The

Gate News reports: according to Gate’s official announcement, Gate ETF will list multiple new 3x ETF assets, including SPCX3L/3S (SpaceX), MRVL3L/3S (Marvell Technology), ANTHROPIC3L/3S (Anthropic), OPENAI3L/3S (OpenAI), and more. Trading will open on June 5, 2026 at 17:00 (UTC+8), with an initial net value of 1 USDT. Gate ETF is a trading product with an automatic rebalancing mechanism. Each day, based on profit and loss, it adjusts the leverage ratio to return to the target leverage multiple.

NVIDIA CEO Huang Jen-Hsuan arrived at Incheon International Airport this afternoon around 1 p.m. and proceeded directly to T1 Base Camp, an esports cafe operated by esports organization T1. He met with five T1 League of Legends starting players, including team captain Lee Sang-hyeok, known as

According to BlockBeats, on June 5, on-chain attack frequency over the past two years has surged to 2.8 times the pre-AI baseline. From 2020 to 2021, monthly on-chain attacks averaged 3.7 incidents, while from June 2024 to June 2026, the monthly average reached 10.4 incidents, totaling 250 attacks.