According to BlockBeats, on August 26th, OKX Flash Earn Lite will launch the AEON "Staking to Earn" program from 15:00 on August 31st to 15:00 on September 5th, 2026 (UTC+8). During the event, users can participate in the subscription by locking up BTC, ETH, OKB, or AEON to share a 4,100,000 AEON airdrop reward. ETH borrowed through flexible lending will not be counted towards the valid subscription amount.

Users can participate in the subscription starting today, and rewards will be calculated from the official start of the event. Participation can be made through the event link or by clicking "Earn Now" at the top of the OKX App's Explore page.

According to Beating AI News, Dylan Patel, founder of SemiAnalysis, predicted in his latest podcast that by 2028, OpenAI and Anthropic may account for 70% to 80% of the world's new AI computing power, with a combined computing power exceeding 100GW. Patel stated that the two companies currently account for approximately 30% of the world's annual new computing power, and this percentage is still rising rapidly.

Patel points out that the business model of cutting-edge AI labs is changing, with a significant increase in the efficiency of AI computing power output. Anthropic currently generates approximately $50 million in revenue per megawatt of computing power, and this could potentially rise to $100 million. This allows OpenAI and Anthropic to purchase or lease computing power at a high price of $25 million to $50 million per megawatt.

More importantly, Patel projects that global AI-related capital expenditures will reach approximately $11 trillion between 2024 and 2029, with over $5 trillion requiring debt financing. Given the potentially much higher returns on AI infrastructure compared to traditional industries, tech giants may accept higher financing costs, thereby pushing up overall credit rates and putting pressure on traditional asset valuations and highly indebted economies.

Furthermore, Patel believes that the additional computing power in the future may not be primarily used to provide model inference services to external parties, but may instead flow more towards internal research and development and model self-improvement within AI labs.

According to Beating AI News, the surging demand for AI computing power is continuously driving up memory chip prices, and upstream cost pressures are beginning to be passed on to consumer electronics and hardware terminals. Recently, Amazon raised prices on several hardware products, with its best-selling entry-level Echo smart speaker increasing from $49.99 to $79.99, an increase of approximately 60%; and the 16GB basic e-reader increasing from $109.99 to $149.99, an increase of approximately 36%.

Before Amazon, tech companies like Apple, Microsoft, and Dell had already addressed rising storage costs by increasing product prices and reducing standard memory in hardware. IDC previously warned that the global memory supply shortage could continue until 2027, and the industry expects memory chip prices to peak in 2028 before gradually declining.