Crypto payment cards processed $759 million in spending across nearly 8.8 million transactions in July 2026, according to fresh data shared by a16z crypto on X. The figures were up sharply from roughly $306 million and about 5.2 million transactions a year earlier, pointing to much heavier use of stablecoin-linked cards for everyday purchases rather than large transfers. Based on the monthly total and transaction count, average ticket size came to about $86. The report said cardholders typically spend crypto assets, mostly stablecoins, which are converted into local fiat at checkout, letting merchants treat the payment like a standard card purchase. Several issuers are active in the segment, including RedotPay, EtherFi, KAST, Karta, Plasma One, Tria and Wirex One, with RedotPay described as the largest contributor by spending volume. Paymentscan data also showed a major shift in stablecoin mix. In July 2026, USDC accounted for 58% of crypto card spending and USDT for 26%, while euro stablecoin EURe had held as much as 88% in early 2024. The story also flagged limits in the dataset: some figures are self-reported by issuers and not fully verifiable on-chain, and activity is concentrated among a small number of providers.

On August 8, 2026, at 19:47 Beijing time, a hacker exploited a liquidity pool vulnerability in Bifrost and stole approximately $720,000 in assets. The affected pools were the vDOT single-asset pool, the vASTR/ASTR pool, and the vMANTA/MANTA pool. According to Bifrost's monitoring, the stolen assets were initially deposited into HitBTC and later transferred to Binance. Bifrost has contacted Binance's security department to request a freeze on the involved funds. Additionally, Bifrost has submitted a report to law enforcement, along with an on-chain evidence package that includes transaction tracing, wallet addresses, and timestamps. In the wake of the incident, Bifrost has paused all liquidity mining rewards and is conducting a comprehensive security review. The total value of the stolen assets was estimated at roughly $720,000, which is based on Bifrost's monitoring data.

Bifrost has reported a security breach in which a hacker exploited liquidity pool vulnerabilities and stole approximately $720,000 worth of assets. According to Bifrost's monitoring, the attacker drained funds from the vDOT single-asset pool, as well as the vASTR/ASTR and vMANTA/MANTA pools. The stolen assets were deposited into HitBTC and later transferred to Binance. Bifrost has contacted Binance's security team to request a fund freeze, and has also submitted a report to law enforcement, including an on-chain evidence package covering transaction tracing, wallet addresses, and timestamps. In the meantime, Bifrost has suspended all liquidity mining rewards and launched a comprehensive security review. No details about the vulnerability's root cause or the identity of the attacker have been disclosed so far.

An OpenAI executive is signaling that the current generation of AI agent tooling may have a short shelf life. Thibault Sottiaux, OpenAI’s general manager for product and platform and the executive cited as overseeing ChatGPT and Codex, said Codex could look like a primitive tool in another two to three months. The remark points to a shift already visible in agent engineering: local, laptop-based harness setups are running into hard limits as models take on longer reasoning, more autonomy, and heavier parallel workloads. The article argues that today’s harness layer — covering context management, tool use, state persistence, environment isolation, and recovery — is being stretched by next-generation demands. It lists three pressure points: constrained local CPU and memory when agents run many subtasks at once, the impracticality of keeping laptops online for jobs that can last hours or days, and the difficulty of handling large-scale context compression, state sync, and centralized logging across many concurrent agents. It also points to early infrastructure responses. Codex already supports asynchronous cloud execution, while providers such as E2B, Daytona, Fly.io, and Modal are building cloud micro-sandbox environments for AI agents. The piece says teams at OpenAI, Anthropic, and Cognition are moving their focus from prompt tuning to system-level harness design, and cites Anthropic’s February demonstration of 16 Claude instances working across 2,000 cloud sessions to build a C compiler.

IOSG argues that the current storage rally is being driven by artificial intelligence, but not in the way traditional IT buyers used to think about storage. In its view, the market is no longer rewarding raw capacity first. It is rewarding the ability to keep GPUs fed, move checkpoints quickly, support retrieval-augmented generation with very low latency, and raise overall compute utilization across tightly coupled infrastructure stacks. That shift, the article says, is why components such as HBM, DRAM, CXL, enterprise SSDs, SSD controllers, NVMe pathways, and performance storage software have become central to the AI investment narrative. The piece draws a sharp distinction between AI storage and decentralized storage. AI storage is framed as an efficiency system built for hot data and commercial output. Decentralized storage, by contrast, is described as a trust system for cold data, focused on permanence, censorship resistance, auditability, and public memory. IOSG uses Filecoin and Arweave as the main examples, outlining how the two networks diverge in architecture and product direction, while also listing persistent problems across the sector, including weak enterprise service layers, retrieval limits, supply-demand incentive mismatches, privacy and compliance tensions, and token economics that can amplify market cycles rather than solve product-market fit.

Monthly spending on crypto payment cards has reached $759 million, according to fresh data from a16z crypto. The figure marks a jump from the niche-scale volume recorded in earlier periods. These cards let users spend cryptocurrency wherever traditional card networks are accepted; at the point of sale, the stablecoin is converted into local currency, so the transaction looks like an ordinary card payment to the merchant. As of July 31, RedotPay accounted for the vast majority of spending across card programs, while EtherFi, KAST and Karta have been gradually expanding their share since late 2025. a16z crypto also said cardholders don't need a traditional bank account: they can deposit stablecoins with the issuer or hold tokens in self-custody. That, the firm says, broadens access to dollar-denominated accounts for global users and gives stablecoin holders a convenient way to spend.

Wall Street turned cautious ahead of the U.S. July nonfarm payrolls report, with the three major stock indexes ending lower, Treasury yields climbing and the dollar testing the 100 level. Oil jumped after Iran advanced a draft rule for transit in the Strait of Hormuz, while LME copper neared historic highs after the Democratic Republic of the Congo moved to ban exports of copper-cobalt concentrates. At the same time, investors reassessed AI-related spending and guidance risk after sharp sell-offs in application software and memory names. The market is also watching a widening split in payroll forecasts, ranging from 18,000 to 83,000 new jobs. That dispersion has raised the stakes for Friday’s data as traders weigh the odds of a September rate hike. St. Louis Fed President Musalem said the probability of inflation staying above target has increased and signaled support for higher rates, while the Financial Times reported that Warsh could also push for a September hike if inflation runs hot in coming weeks. Within equities, SpaceX and several space-linked names outperformed, while Western Digital, SanDisk, AppLovin and Datadog were hit hard after earnings-related disappointments. Alphabet’s $25 billion bond sale, backed by roughly $115 billion in orders, added to supply pressure in long-dated Treasuries and sharpened concerns over how AI infrastructure spending may weigh on free cash flow.

Binance said in an official announcement on Aug 5 that Cross Margin, the Unified Account mode and Unified Account Pro will accept 10 bStocks tokens as eligible collateral starting at 20:00 UTC+8 on Aug 5, 2026. The newly supported tokens are BitMine Immersion Technologies (BMNRB), Super Micro Computer (SMCIB), IREN Limited (IRENB), ASML (ASMLB), Netflix (NFLXB), AST SpaceMobile (ASTSB), Coherent (COHRB), Credo Technology (CRDOB), USA Rare Earth (USARB) and Astera Labs (ALABB). Corresponding bStocks trading pairs will also have leveraged trading enabled. Eligible users will be able to put these bStocks tokens to use as margin collateral. That widens the mix of collateral options available for margin trading on the platform. The change covers the cross margin account, the unified account mode and the unified account pro version, and all ten tokens are scheduled to become eligible at the same time: 20:00 UTC+8 on Aug 5, 2026. BlockBeats reported the update, citing Binance's official announcement.