Coinbase CEO responds to stablecoin yield controversy: Fundamentally distinct from bank interest, so they do not need to be subject to bank capital and liquidity requirements.

Insight Beating AI News Flash: Veteran chip designer and hardware analyst Mr. Bubble stated in an interview on the Frictionless Podcast that the core bottleneck of AI computing infrastructure is shifting from simply boosting chip computing power to advanced packaging, system interconnection latency, and storage tier management. Mr. Bubble noted that the rapid rise in costs of advanced process nodes and the limited growth in transistor density have put traditional Moore’s Law under economic challenges. Instead of continuing to shrink process nodes, the industry is increasingly using advanced packaging to connect multiple chips into a single computing system. He pointed out that the basic unit of AI computing power in the future may no longer be a single chip or server, but a complete rack or even multiple racks, with PCB, packaging, and interconnection capabilities emerging as key limiting factors. On the storage front, he argued that as AI inference context windows expand, massive historical context data should not all occupy expensive High Bandwidth Memory (HBM). The industry may increasingly adopt flash memory offloading, keeping high-frequency data in HBM while migrating low-frequency context to larger-capacity, lower-cost storage layers. He further projected that 3D DRAM could become a key direction for next-generation memory technology. Regarding AI infrastructure investment, Mr. Bubble believes that as prefill and decoding gradually adopt a decoupled architecture, the importance of system-level hardware design will further rise. He also noted that compared to directly betting on large language model companies, enterprises with hardware, packaging, storage, and other key infrastructure capabilities may play a more long-term role in the AI industry chain.

Insight Beating AI News Flash: Meta’s personal AI agent Muse has received diametrically opposite treatment on Amazon and Shopify. Amazon has blocked Muse from searching for products and making purchases on behalf of users on Amazon.com, with users seeing a direct prompt stating “Unauthorized AI Agent” violating terms of service. Amazon noted that Meta did not obtain prior permission, and Muse also fails to proactively disclose its identity as an AI agent when browsing the site. Around the same time, Shopify announced a deep partnership with Muse, integrating Shop Pay into all its stores. Shop Pay is essentially Shopify’s one-click checkout solution, allowing users to complete purchases directly via Muse after selecting products, without needing to re-enter shipping and payment details on the merchant’s website. The core difference between the two platforms’ stances stems from their distinct business models. Amazon operates as a shopping gateway itself: users typically search, compare, view recommendations, and finalize purchases on Amazon, and external agents like Muse could siphon off this entire user journey. Shopify, meanwhile, primarily provides product, order, and checkout infrastructure for independent merchants. It does not matter whether consumers come via Muse, ChatGPT, or Google—so long as they ultimately make a purchase from a Shopify merchant, Shopify gains an additional sales channel. For Amazon, Muse is potentially siphoning its core traffic; for Shopify, Muse acts more as a customer acquisition driver.

According to Farside's monitoring, U.S. spot Ethereum ETFs recorded a net inflow of $270 million yesterday, marking two consecutive trading days of large net inflows.

Cardano has been integrated into the official x402 software suite, enabling developers to build applications and AI agents that support payments in ADA and Cardano network tokens, allowing AI agents to complete on-chain payments directly for services like online data and computing resources. Launched by Coinbase in 2025, x402 converts the HTTP "402 Payment Required" status code into a native payment mechanism: service providers return pricing and payment instructions, AI agents sign transactions and access data or computing services after payment verification, eliminating the need for manual account registration, bank card details input, or subscriptions. Cardano Foundation engineers finalized the relevant specifications in June this year, and developed the client, server, and Facilitator component responsible for transaction verification and submission. The first version currently supports TypeScript, with a Python version planned for future release. However, the Facilitator has only completed actual transaction tests on Cardano’s pre-production network and has not yet been deployed on the mainnet, so it cannot yet be confirmed that ADA has achieved large-scale commercial AI agent payments. Solana, the XRP Ledger, and multiple Ethereum-compatible networks already support x402, and Cardano is joining the competition in AI agent machine payment infrastructure.

Beating AI News: JPMorgan Chase CEO Jamie Dimon stated that the global "hyperscale cloud vendor ecosystem" – including large cloud computing firms and their upstream and downstream suppliers – is expected to spend approximately $700 billion on AI infrastructure this year, more than doubling from around $300 billion last year, and could further exceed $1 trillion by 2027. Dimon noted this round of AI investment is expanding into data centers, chips, servers, power, and industrial infrastructure, with annual new investment equivalent to roughly 1% of U.S. GDP, which may drive inflation in the short term. As AI data center construction accelerates, infrastructure such as land, power, transformers, gas turbines, and transmission lines has also become new investment priorities. He also pointed out that not all AI investments will generate clear returns in the short term; some outlays are closer to costs tech giants must bear to maintain their competitive positions. The market is increasingly focused on whether, as AI infrastructure investment moves toward the trillion-dollar scale, related revenues and cash flows can cover the continuous expansion of capital expenditures.

According to HTX market data, established meme coins are seeing a broad rally, with the following 24-hour gains: PEPE up over 32%, DOGE up over 18%, FLOKI up over 15%, WIF up over 27%, SHIB up over 12%, and BONK up over 16%.