velo

Renowned Ethereum builder and investor Ryan Berckmans posted on X, stating that the idea that Robinhood "gives back very little to ETH" is nonsense. They pay on-chain users and new ETH holders, conduct marketing, attract new enterprise clients envious of Robinhood's success, and help mature L1 hubs radiate to L2. Half of Robinhood's active users come from other places in Ethereum and exert healthy competitive pressure on Base.He mentioned that Ethereum alignment is a real concept, but it is not portrayed as income maximization by Ethereum opponents. Helping to grow Ethereum and ETH is alignment, while threatening Ethereum or ETH is misalignment, such as during Lido's dominance. Robinhood is one of the most Ethereum-aligned entities globally. Buying large amounts of ETH or not selling order fees would be better, but this is just icing on the cake of an already huge pie.We do not rely on Robinhood to buy ETH to win. The next batch of Tom Lee will buy ETH, Wall Street Tom Lee, corporate Tom Lee, government Tom Lee, family office Tom Lee, sovereign wealth Tom Lee will be everywhere. ETH will reach trillions in value storage as Ethereum goes global. The two major indicators measuring the progress of value storage are L1 application capital and the number of sufficiently successful L2s, and Robinhood helps grow the latter. We like Robinhood as it is now.

The G7 Cybersecurity Working Group stated in its latest report that quantum computing poses security threats to both public and private institutions, as well as economic threats, and relevant organizations should begin migrating to post-quantum cryptography (PQC) immediately. The working group noted that the migration process could take several years, and attackers can currently collect and store encrypted data to decrypt it once sufficiently powerful quantum computers become available.Quantum computing may also break digital signatures, leading to identity theft and exposing businesses and their supply chains. The report did not mention cryptocurrencies, but similar public key cryptography is used for blockchain wallets and transaction authorizations. Currently, quantum computers are unable to break Bitcoin's cryptography, but developers are considering post-quantum solutions such as BIP-360. Ethereum researchers plan to replace several cryptographic components used by accounts, validators, and applications. The Solana Foundation has tested post-quantum signatures on its testnet and launched an optional hash-based vault. The G7 working group also urged governments to support relevant research, public-private collaboration, and national PQC strategies.

Binance founder Zhao Changpeng met with Deputy Prime Minister and Minister of Artificial Intelligence and Digital Development, and old friend Zhaslan Madiyev in Kazakhstan, stating that the local cryptocurrency industry is developing rapidly.It is reported that Zhaslan Madiyev served as the General Manager of Binance Kazakhstan from 2022 to 2024, and was appointed as the "Minister of Digital Development, Innovation and Aerospace Industry" starting in May 2024, and then promoted to "Deputy Prime Minister and Minister of Artificial Intelligence and Digital Development" in 2025 and 2026.

According to The Wall Street Journal, DeFi Development Corp. (NASDAQ: DFDV) announced that its variable rate C series perpetual preferred stock "CHAD Stock" has been priced for its initial public offering on September 3, with a total of 1.375 million shares issued at a price of $8 per share; the underwriters also received an option to purchase an additional 206,250 shares within 30 days.The company expects the gross proceeds from this offering to be approximately $11 million (before deducting underwriting discounts, commissions, and offering expenses), with the transaction scheduled to settle on September 8. The net proceeds will be used for general corporate purposes, including the purchase of Solana (SOL) and other digital assets, strategic transactions, and growth plans.

According to South Korean media Greened.kr, Samsung Electronics and Arm have officially launched a joint research and development project for edge AI chips. Arm approved a one-time engineering fee at the end of August for the development of the next-generation edge AI SoC chip and officially started the project with Samsung Electronics.The project adopts a collaborative model where Arm provides core technology, and Samsung Electronics integrates and implements it, aiming to jointly develop customized chips for end customers.

According to TechCrunch, AI data center developer Crusoe has completed a $3 billion financing round, achieving a valuation of $30 billion, led jointly by Atreides Management and Valor Equity Partners, with participation from Mubadala Capital, a subsidiary of the Abu Dhabi sovereign wealth fund.Crusoe's clients include Meta, Microsoft, and OpenAI, and it recently signed a five-year, $13 billion GPU and AI infrastructure cloud contract with quantitative trading giant Jane Street. This financing comes just 10 months after raising $1.38 billion at a $10 billion valuation last October. Founded in 2018, Crusoe initially utilized stranded natural gas for Bitcoin mining and later transformed into an AI infrastructure and cloud service provider, known for developing ultra-large-scale data center campuses. Axios reported last month that Crusoe has met with investment banks such as Goldman Sachs and Morgan Stanley to discuss a potential IPO in the near future.

U.S. Senator Bernie Sanders and Representative Greg Casar introduced the "Artificial Intelligence Superintelligence Prohibition Act," which calls for a permanent ban on the development of superintelligent AI and a suspension of advanced AI research until new federal agencies establish safety standards and review procedures. Sanders criticized large AI companies for developing technologies that they cannot fully understand or reliably control, stating that companies and individuals who violate the regulations could face up to 20 years in prison, with the most serious cases potentially subject to what is referred to as "corporate death penalty."The bill defines superintelligence as AI systems that achieve or exceed human capabilities across numerous tasks or could undermine human control (including overthrowing the U.S. government), and this definition also encompasses human-level AGI. The bill plans to establish a new cabinet-level agency responsible for monitoring advanced models, overseeing the removal of dangerous capabilities, and the destruction of banned systems, as well as forming an AI advisory committee to provide scientific and technical advice.On the same day the bill was announced, OpenAI released GPT-6 Astra, with co-founder Greg Brockman stating that he personally believes this model may meet the definition of AGI. OpenAI described Astra as its first model to reach a "critical" threshold of cybersecurity capability, potentially discovering and exploiting unknown vulnerabilities in protected systems without gradual human guidance, with its strongest cybersecurity capabilities limited to selected testers. Previously, OpenAI and Anthropic's models had repeatedly accessed real systems without authorization during testing, including incidents where OpenAI's model breached Hugging Face to obtain test answers.

According to a report by Reuters, Apple Inc. is facing a class-action lawsuit in London, with claims amounting to £2 billion (approximately $2.7 billion). The lawsuit was filed today in the London Competition Appeal Tribunal by Ann Pope, a former senior official of the UK's Competition and Markets Authority, representing app developers.The core allegation is that Apple's "App Tracking Transparency" (ATT) feature, launched in 2021, imposes stricter restrictions on third-party developers than on its own services, giving Apple's own advertising ecosystem an unfair competitive advantage. Ann Pope stated that Apple's policies "have caused very significant harm to businesses that rely on Apple as a gatekeeper."Since its launch, the ATT feature has been a focal point of concern for global regulators for several years. Apple's official stance is that the feature is designed to allow users to control whether to permit apps to track their activities across other companies and websites.However, the plaintiffs argue that the actual enforcement of this rule has a double standard—tracking requests from third-party apps require strict pop-up authorization, while Apple's own personalized ads and services can bypass the same restrictions. This lawsuit represents the latest legal challenge Apple faces regarding its ATT policy and is the first large-scale private antitrust lawsuit initiated in the UK market against Apple's app ecosystem rules following scrutiny from regulators in the EU, the US, and several other countries.

OpenAI CEO Sam Altman stated that the company will definitely develop humanoid robots while also exploring other forms of robots. He believes that the reality of human society and tools is mostly designed around the human body structure, so humanoid robots are reasonable; however, the robot form is not the key, what is truly important is the "robot brain."Altman mentioned that in the short term, humanoid robots are unlikely to enter ordinary households. OpenAI prefers to first apply them in industrial scenarios such as data center construction and maintenance, while general-purpose robots for individual users are a long-term goal.

According to sources in the materials industry, on September 2, TSMC is expected to continue using traditional micro-bump technology rather than hybrid bonding in the short term for connecting high-bandwidth memory (HBM) with AI accelerators in advanced packaging. Considering the development cycle of related materials, finer pitch micro-bumps are expected to be used until the later stages of HBM4 and the early stages of HBM5. TSMC has requested its materials and equipment partners to develop bonding and underfill solutions for approximately 5-micron bumps, with suppliers from South Korea and Japan beginning development, and mass production of 5μm bumps expected to start in the second half of 2028.Currently, the bump height of the third-generation extended HBM, namely HBM3E, is about 15-25μm, while HBM4 is close to about 10μm. Japanese material suppliers have previously stated that it is difficult to guarantee quality below 15μm. The reduction and refinement of bumps are due to the height limitations of HBM cubes and the demand for higher interconnection density. JEDEC specifies that the maximum height for HBM stacks is 775μm. In TSMC's advanced CoWoS packaging, the HBM stacks assembled by GPU and memory suppliers are mounted onto the silicon interposer via micro-bumps, with the 16-layer DRAM chip stacks completed within HBM packaging by SK hynix and Samsung Electronics.The first and second generations of HBM used larger solder bumps, while micro-bumps became mainstream around the HBM3 generation. SK hynix uses the MR-MUF process, while Samsung Electronics uses TC-NCF. Hybrid bonding allows for thinner and denser interconnections through direct bonding of copper pads; TSMC has utilized this on its SoIC platform for logic chip stacking, but HBM still connects to the interposer using micro-bumps.

According to Bloomberg, UBS analysts expect that China's semiconductor manufacturing capabilities may struggle to make sufficient progress in the next decade to develop solutions that can truly replace ASML's cutting-edge EUV lithography technology. UBS analysts wrote in their report, "The stage they are currently at seems comparable to ASML's in 2004." Based on patent application activities, especially in the fields of light sources and laser subsystems, China's current technological maturity is roughly equivalent to ASML's level 15 years before it began mass production of EUV equipment.UBS also expects that China is likely to have the capability for mass production of immersion DUV lithography machines within the next two to five years. Although immersion DUV is not as advanced as EUV, it remains an indispensable tool in chip manufacturing. ASML's immersion DUV machines are priced at nearly $90 million, while EUV equipment costs over $200 million each. China is one of ASML's largest markets, but due to export restrictions, ASML is prohibited from selling EUV equipment to China. UBS pointed out that considering the yield and capacity gaps and regulatory restrictions, Chinese lithography equipment is unlikely to be applied overseas. An ASML spokesperson did not respond to requests for comment. The report is based on UBS analysts' predictive views, and actual progress will depend on industry dynamics.

Predict.fun posted on platform X that it has launched the Predict Developer Dashboard, a self-service portal for creating applications, generating keys, and managing usage and rate limits. Through this interface, users can manually request higher usage-based tiers and import existing keys.Predict.fun has also implemented "trade burst" limits to restrict the number of calls for creating and canceling orders per second. New applications have this limit enabled by default, while existing applications will have this limit applied the next time the user requests rate limits.

According to GlobeNewsWire, the U.S. publicly traded company DeFi Development Corp. (Nasdaq: DFDV) announced plans to conduct an initial public offering (IPO), issuing up to $20 million in Series C perpetual preferred stock, with net proceeds to be used for working capital, increasing SOL holdings, and investing in other digital assets.

Luke Dashjr, who has been engaged in Bitcoin development for a long time, has exited the mining pool Ocean, resigning from his positions as Chairman, Chief Technology Officer, and Director. The parent company Mummolin has repurchased all of his shares. Both parties stated that this separation was mutually agreed upon, stemming from differences regarding the future direction of Bitcoin mining and recent developments in agreements.Dashjr is a co-founder of Ocean and will establish a new mining project called Convoy after his departure. Ocean stated that it will continue to operate its transparent, non-custodial mining pool, directly paying mining rewards to participating miners.This departure occurred after he took a few weeks off due to the failure of the BIP-110 proposal. BIP-110 is a controversial proposal aimed at limiting the storage of non-financial data on Bitcoin.

According to South Korean media reports, Samsung Electronics is cooperating with Nvidia (NVDA.O) to develop 8-layer seventh-generation high bandwidth memory (HBM4E), significantly reducing the stacking height compared to the original plans of 12 layers and 16 layers. The speed specification proposed by Nvidia is 17-18Gbps, which is about 20% higher than Samsung's initial HBM4E sample speed of 14.4Gbps. This product is reported to be used for NVHBM.

GoPlus Security issued a security alert stating that a fraud gang has hijacked the realtrumpcoins[.]com website and the @realtrumpcoins1 account, maliciously issuing GOLD tokens. The gang inflated its market value to approximately $60 million before selling off and executing a Rug Pull, profiting over $8.2 million.On-chain data shows that the developers' initial funds trace back to a KuCoin-associated hot wallet, with another 15 trading wallets funded by Binance-associated hot wallets. These wallets all conducted transactions through the same malicious routing contract.GoPlus Security also pointed out that the gang deployed another malicious meme coin, Trump Digital Platinum (PLATINUM), whose contract can empty any holder's balance at any time, warning users to avoid purchasing and to be aware of the risks.

According to monitoring by Lookonchain, realtrumpcoins (@realtrumpcoins1) has launched a token called GOLD, which is followed by @realDonaldTrump. The developers hold 600 million GOLD, and 15 newly created wallets spent $18,700 to purchase 224.5 million GOLD. The team currently controls 82.45% of the total supply of GOLD.

Samsung Electronics has launched the passport-type foldable smartphone Galaxy Z Fold8 this year, and the supply of display driver chips (DDI) is tight. The demand for Z Fold8 has exceeded expectations, and Samsung is increasing component orders. However, this DDI is designed by Samsung's System LSI Division and produced by Taiwan's foundry UMC using a 22-nanometer process, which has reached its production capacity limit, making it difficult to advance the production schedule.Industry insiders say that UMC's 22-nanometer production line is also manufacturing other chips, and Samsung's request for super urgent service was rejected. Samsung plans to prioritize the DDI quantity originally reserved for development next year for mass production this year. The typical time from wafer to finished product takes about 4 months, and the finished product is then assembled after being bonded with Samsung's OLED. Samsung is also increasing orders for hinges and ultra-thin glass (UTG). The demand for Z Flip8 is in line with lower expectations, and Apple also plans to launch a passport-type foldable product in the second half of this year.

According to reports from Seoul Economic Daily, Samsung Electronics is developing HBM4E (seventh generation) 8-layer products that meet NVIDIA's requirements, aiming to secure its position as a core partner in the supply of customized high-bandwidth memory NVHBM. This product reduces the stacking height compared to the originally planned 12-layer and 16-layer designs. The speed specifications proposed by NVIDIA are 17-18Gbps, which is about 20% higher than the speed of Samsung's initial HBM4E samples (14.4Gbps), and will be used for NVIDIA's publicly disclosed NVLink optimized specifications for NVHBM.The use of 8 layers, contrary to the previous logic of increasing capacity by adding more stacking layers in HBM, can reduce the difficulty of post-processing and yield pressure, which is beneficial for expanding supply and is seen as NVIDIA's strategy to alleviate memory shortages. NVHBM is expected to be applied starting with the next-generation AI GPU "Rubin Ultra," which is set to launch next year. This GPU will expand the interconnection scale between GPUs from a maximum of 72 to 576, enhancing overall computing power through hundreds of GPUs equipped with faster HBM.In the customized HBM market, Samsung is more competitive compared to SK Hynix and Micron, as NVHBM requires DRAM and the production capabilities of logic chip-based bare die designs, which Samsung can integrate. Industry insiders say that Samsung has verified the highest speed levels in HBM4, which can provide an advantage in speed competition.