Strategy Inc (MSTR) strengthened its balance sheet after lifting its Strategy cash reserve to $3.75 billion, following a $544.5 million equity raise through its at-the-market stock program. The company kept its Strategy Bitcoin holdings unchanged at 843,775 BTC, signaling a continued focus on liquidity while preserving its long-term Bitcoin strategy. Shares of MSTR and preferred stock STRC both traded higher in premarket activity.

Strategy Cash Reserve Expands Through New Equity Offering
The Strategy cash reserve increased by $525 million from the previous week after the company sold more than 5.4 million Class A shares. According to a filing with the U.S. Securities and Exchange Commission, the offering generated $544.5 million in net proceeds.

Executive Chairman Michael Saylor said the Strategy cash reserve is sufficient to cover roughly 2.1 years of preferred stock dividend payments. The stronger liquidity position gives the company greater financial flexibility without requiring sales of its Bitcoin treasury.
Strategy Cash Reserve Supports Long Term Bitcoin Strategy

Alongside the capital raise, Strategy repurchased 288,930 STRC preferred shares for $25 million while leaving its Bitcoin position untouched. The company has now gone three consecutive weeks without buying or selling Bitcoin, maintaining 843,775 BTC acquired at an average purchase price of $75,476 per coin.

The Strategy cash reserve also supports interest obligations and future capital management initiatives. Strategy still has $975 million authorized for preferred share repurchases and $1 billion available for common stock buybacks.

Meanwhile, Michael Saylor sparked debate after arguing that Bitcoin’s broader adoption depends on integration with traditional banks. His comments divided the Bitcoin community, with critics pointing to Bitcoin’s original peer-to-peer vision, while supporters argued institutional participation could accelerate mainstream adoption.

Samsung Electronics and Broadcom have entered into a memorandum of understanding (MOU) in order to strengthen their collaboration in the fields of memory, foundry manufacturing and advanced packaging. They expect to spend more than $200 billion as a result of their partnership till 2030. The agreement comes at a time when many AI chipmakers are trying to secure their supplies of high-bandwidth memory (HBM), one of the key challenges in the industry.

The agreement was made public at an AI summit held in San Francisco where South Korean President Lee Jae Myung and representatives of prominent AI firms took part. The South Korean presidential office regarded the Broadcom collaboration as one of the initiatives to enhance the local AI semiconductor industry.
While Samsung acknowledged the anticipated financial projections and scale of the partnership, the firm and the President’s office of South Korea have declined to specify how many billions they would supply each year, what the prices would be, when shipments would be scheduled, or how the amount would be split across memory, foundry, and advanced packaging. Due to the lack of any order or buying agreement, the estimated value of $200 billion would represent a mere long-term strategy, rather than already determined revenues.

Why Broadcom buying memory matters beyond the two companies
Broadcom does not manufacture memory chips, but it has become one of the world’s largest designers of custom AI accelerators. Samsung said the broader partnership will support Broadcom’s next-generation AI chips through advances in HBM, foundry manufacturing and advanced packaging. Broadcom already develops Google’s Tensor Processing Units (TPUs) and Meta’s MTIA accelerators, both of which require large amounts of HBM.
This is the reason why the agreement has the potential to create waves in the AI sector. HBM continues to be one of the most constrained components in terms of availability in AI hardware. The Samsung-Broadcom deal can secure an entire manufacturing capability that other chip developers may want to access as well.

It is already clear that demand is on the rise. According to Seoul Economic Daily, SK Group Chairman Chey Tae-won has said Broadcom has been continuously asking for what he calls an astonishing amount of memory chips, emphasizing the way hyperscale AI clients are working hard to make sure they are able to buy enough supplies for the future.
Data from the industry supports this claim. TrendForce’s first quarter 2026 report on HBM technology showed SK hynix still firmly at the head of the market, while Samsung registered the fastest recovery due to increasing HBM3E shipments followed by the launch of HBM4 commercialization. An update from TrendForce, issued later, reported that Samsung had succeeded in validating HBM4 and started shipping it ahead of any other supplier, which reinforces its position against a rising demand for AI.
The deal arrives as memory supplies remain tight
Timing may be just as crucial as the agreement.

The cost of memory has experienced growth over the course of the year, given the high demand for AI servers which exceeds the supply of such servers, and experts do not expect this situation to change any time soon.

On July 3, the forecasting company TrendForce predicted that the contracts for DRAM chips in the third quarter of 2026 would rise by 13-18% compared to the previous quarter, while NAND flash chip prices are expected to hop by 10-15%, stressing the fact that the market for DRAM chips is very tight.
According to SemiAnalysis, the industry has entered a so-called “silicon shortage phase,” during which time production of advanced logic and memory is unable to keep pace with the funding of AI infrastructure. In this scenario, firms that are prepared to make long-term commitments secure more certainty with regard to future supply while manufacturers find themselves locked into long-term contracts with customers for their extremely limited capacities.
Samsung’s bid to regain ground in AI memory
Samsung formed the agreement with Broadcom amidst efforts to bolster market position in AI memory after losing a lead in HBM to SK hynix. The company stated that the partnership with Broadcom brings together its expertise in memory, foundry production, and advanced packaging to create a comprehensive platform for AI customers.

Samsung had revealed a few months ago that it had started mass production of HBM4, and then, on May 29, released what it claimed was the first 12-layer HBM4E samples available for large customers. Samsung asserted that the latest addition to its memory lineup could achieve transfer speeds of 16 gigabits per second, as well as 3.6 terabytes per second bandwidth per stack, targeting future AI accelerators.
Its foundry business has also gained momentum. SemiAnalysis reported that Samsung Foundry has secured Tesla’s AI5 and AI6 programs alongside TSMC and has entered Nvidia’s data-center supply chain.

If the Broadcom memorandum eventually turns into firm production orders, Samsung would add another marquee AI customer to its expanding semiconductor portfolio. The South Korean Presidential Office also said Korean and global technology companies announced about $950 billion in semiconductor cooperation during the AI Summit, making the Broadcom agreement one of the event’s largest strategic commitments.

Circle, the issuer of USD Coin (USDC), has announced two separate partnership agreements in South Korea, signing memorandums of understanding with Kakao Group and Toss Bank to explore stablecoin-based payment infrastructure in the country.

The agreement with Kakao Group will focus on exploring blockchain-based payment infrastructure in South Korea. Kakao operates a wide range of consumer services including KakaoTalk, Kakao Pay, and KakaoBank, making it one of the country's most widely used technology platforms.
Circle and Toss Bank Assess Stablecoin Payment Opportunities
Circle's second partnership, signed with internet-only lender Toss Bank, centers on assessing opportunities for stablecoin-based payments. Toss Bank has been broadening its presence in the cryptocurrency sector and announced a partnership with the Solana Foundation in June 2026 to develop blockchain-based financial infrastructure for global users.
The two new agreements follow a pattern Circle has been building in South Korea over the past several months. In April 2026, Circle signed agreements with Upbit and Bithumb, the two largest cryptocurrency exchanges in the country, which together regularly account for more than 95% of daily crypto trading volume in South Korea.

Circle CEO Points to Broader Technology Collaboration
At the time of the Upbit and Bithumb deals, Circle CEO Jeremy Allaire said the partnerships covered continued work on the promotion and adoption of USDC on Korean exchanges, alongside other technology collaborations tied to Circle's broader business activities.

As of July 23, USDC had a total circulating supply of $74.4 billion, meanwhile Tether's USDT, the largest stablecoin by supply, had a circulating supply of $184.3 billion.

Franklin Templeton, a leading global asset management firm overseeing over $1.5 trillion in assets, has identified artificial intelligence agents as the next major growth area for blockchain and cryptocurrency. Sandy Kaul, the firm’s head of digital assets and innovation, outlined this vision in a recent post on X.

AI agents and blockchain infrastructure
According to Kaul, the emergence of an AI-driven agent economy will generate significant demand for blockchain protocols capable of supporting machine-to-machine micropayments. She noted that legacy payment networks, including widely used card systems, face challenges meeting the fast and low-cost requirements of automated digital agents.

Kaul highlighted the limitations of established card networks, with fees and settlement speeds unsuited for the high-frequency, micro-level transactions typical of automated AI agents. In her view, most investors today focus on acquiring shares of companies aligned with the AI sector, but she questioned whether this approach will remain effective as agentic AI becomes prevalent.

Most investors today buy shares of AI-aligned companies to access the AI growth opportunity, but it remains uncertain if that strategy will hold as agentic AI evolves.

She pointed to blockchain networks including Aptos, Solana, and BNB Chain as well-positioned for this new digital landscape. These platforms can settle transactions within seconds, offering a sharp contrast to the one-to-three business day settlement times seen in systems like the Visa network.
Payment industry leaders have recently examined this topic as well. Payments giant Visa and research platform Artemis published a joint report last week, concluding that traditional cards—designed for infrequent, human-driven transactions—are not adequate for the needs of AI agents. They argued that to support agentic micropayments on a commercial scale, networks require both instant settlement and minimal fees.

Mini dictionary: Agentic economy, a digital ecosystem where AI agents autonomously perform transactions or tasks, often interacting with other machines, users, or protocols without direct human intervention.

Industry response and adoption trends
Within the past few months, several major players have launched tools targeting the intersection of AI and payments. Visa’s crypto division and Tempo, supported by Stripe, both unveiled AI-driven solutions in March. Visa’s new function grants AI agents the ability to process same-day payments.

Meanwhile, new protocols facilitating machine payments are seeing early signs of traction. The x402 payment protocol, a system created by Coinbase, reportedly processed $15 million in adjusted volume through over 109 million adjusted transactions since its introduction in May 2025, according to the joint analysis by Visa and Artemis.

Visa launched its machine-to-machine payments tool to strengthen its presence as the pace of agentic transactions accelerates, while adoption data from Coinbase illustrates practical engagement with the technology in live environments. This suggests interest is building in infrastructure that can support the complex and rapid settlement needs of AI-driven economies.

Visa and Artemis found that traditional payment cards are not built for the frequency or scale required by agentic AI transactions, reinforcing the shift toward blockchain alternatives.

The increasing experimentation and support for agentic AI payments by both blockchain networks and major payment industry companies reflect a growing recognition of the role digital assets may play in powering next-generation automated commerce.

TeraWulf reported that revenue from high-performance computing and artificial intelligence infrastructure significantly exceeded its Bitcoin mining income during the first quarter. The company continues shifting its infrastructure strategy away from traditional Bitcoin mining toward long-term computing contracts.

TeraWulf expands AI computing revenue
TeraWulf generated total first-quarter revenue of $34 million. High-performance computing services accounted for approximately $21 million of that amount.

Meanwhile, Bitcoin mining revenue fell to around $13 million compared with $34.4 million during the same period last year. The report marks one of the clearest examples of AI-related revenue overtaking mining income among public mining companies.

As of March 31, TeraWulf reported total assets valued at approximately $7 billion. However, total liabilities exceeded assets by roughly $77.6 million.

The company also held approximately $3.1 billion in unrestricted cash and cash equivalents.

TeraWulf posted a net loss of $427.6 million during the quarter, compared with a $59.6 million loss one year earlier.

Company reduces reliance on Bitcoin mining
The larger quarterly loss was primarily driven by higher operating expenses, depreciation costs, asset write-downs and increased interest payments.

At the same time, TeraWulf continues restructuring its business around artificial intelligence infrastructure and high-performance computing services.

Last August, the company signed a 10-year agreement with Fluidstack, a cloud platform backed by Google. The agreement covers approximately 250MW of computing capacity with a total contract value of $3.7 billion.

In addition, the first 60MW phase of the Lake Mariner data center project has already begun generating substantial revenue under a separate agreement with Core42.

TeraWulf’s results reinforce the broader transformation underway across the Bitcoin mining industry. Large operators increasingly view AI infrastructure as a more stable and scalable revenue source.

High-performance computing contracts provide predictable long-term cash flow compared with Bitcoin mining, where profitability depends on BTC prices, network difficulty and electricity costs.

As a result, mining companies continue upgrading data centers and reallocating infrastructure toward artificial intelligence workloads.

Analysts believe AI infrastructure could become the primary growth driver for many public mining firms over the next several years.

Industry shifts toward long-term computing contracts
TeraWulf’s report highlights how the mining sector is evolving into a broader computing infrastructure industry.

Mining companies are increasingly using their existing power assets and data center facilities not only for cryptocurrency mining, but also for artificial intelligence and enterprise computing services.

The transition is reshaping the business model of the sector. More operators are prioritizing long-term infrastructure contracts instead of relying exclusively on Bitcoin mining economics.

In Ethereum news today, spot ETFs recorded $105M in net inflows during the week of July 13–17, 2026, the strongest weekly figure since April and a measurable acceleration from the prior week’s $84M.

The two consecutive positive weeks end an eight-week outflow streak, raising the direct question of whether this is a durable institutional re-engagement or a short-term technical bounce that will stall at the first sign of ETH price weakness.

Flow-tracking platforms CoinGlass and Farside Investors both confirm the reversal, with data showing renewed net creations across the Ethereum ETF complex after two months of persistent redemptions.

The prior week’s $84M was itself notable as the streak-breaker; the follow-through to $105M adds weight to the argument that the reversal has legs rather than being a single-week anomaly.

BlackRock’s iShares Ethereum Trust, trading under the ticker ETHA, has consistently accounted for the majority of daily net positive flows across the entire Ethereum ETF landscape. On July 15 alone, in a single session during the reported week, ETHA captured a substantial share of the day’s $53.83M in complex-wide inflows, according to data from BingX.

The concentration dynamic cuts both ways. ETHA’s brand, distribution reach, and institutional trust give it a structural pull that smaller issuers cannot easily replicate, which explains why BlackRock’s ETF inflows have driven ETH price action more than any other single product in the category.

But it also means the health of the entire spot ETF complex is effectively contingent on one fund; if ETHA flows stall or reverse, the broader category tips back into net outflow territory almost immediately.

As of mid-July, cumulative complex-wide net inflows across nine issuers since the category launched in July 2024 totaled approximately $11.07Bn, with total net asset value near $10.4Bn, per BingX data. ETHA’s own cumulative net inflow reached $11.28Bn.

ETH price traded at approximately $1,845 during the inflow week, with the $1,800–$1,900 range serving as a critical demand zone. Buyers have consistently stepped in near the lower end of that band, and the structural logic is straightforward.

Sustained spot ETF inflows create a mechanical bid, because each new creation requires the fund to purchase actual ETH to back its shares. At $80–105M in weekly inflows, this represents consistent buy pressure absent during the eight-week redemption period.

The $1,800 level is therefore not just a technical support reading; it is partially a function of ETF flow dynamics. A breakdown below that zone would likely signal either a sharp deceleration in institutional demand or net outflow resumption, both of which would remove the mechanical bid that has been supporting prices.

The inverse is also true: sustained weekly inflows in the $80–105M range provide a floor that did not exist during the prior two-month drawdown.

Institutional Crypto Momentum and What the Data Confirms and What It Doesn’t
In other Ethereum news, the $105M weekly figure is the best since April, but it remains modest compared with the peaks the category reached during more euphoric periods in 2024 and early 2025. Calling this an institutional comeback is accurate as a directional statement; framing it as a full-scale rotation back into institutional crypto ETF allocation requires more evidence. Two consecutive positive weeks after a prolonged outflow streak is a reversal, not yet a trend.

What the data does confirm is that institutional interest in Ethereum as a strategic portfolio asset has not evaporated, despite two months of redemptions suggesting otherwise. The speed of the reversal, from outflow-heavy weeks to back-to-back inflow weeks accelerating from $84M to $105M, indicates that allocators were watching specific price and macro conditions before re-engaging, rather than abandoning the category entirely.

For context on how Ethereum’s ETF recovery compares to the broader spot ETF landscape, XRP ETF flows have shown a different pattern, which underlines that the current Ethereum inflow momentum is asset-specific rather than a broad crypto ETF tide lifting all products.

The forward scenario is binary and relatively clean. If ETHA sustains its pace of flow through late July and ETH holds the $1,800 support zone, the two-week reversal will validate the start of a genuine institutional re-accumulation phase.

If flows decelerate sharply or ETHA specifically turns negative, the outflow streak resumes and the $1,800 floor loses its structural underpinning. The next two weeks of weekly flow data from CoinGlass and Farside Investors will settle that question more definitively than any price chart reading alone.

Talks remain preliminary.

The proposed agreement could reach $10 billion over two years.
A deal would give Anthropic more capacity while helping Meta seek returns from its expanding AI infrastructure.

Meta Anthropic Talks
Media reported on Jul. 17 that Anthropic could lease computing power from Meta’s data centers, although the companies may not reach an agreement. Reuters confirmed the discussions. The contract would mark a shift for the owner of Facebook and Instagram, which still earns most of its revenue from advertising.
Meta has considered selling access to its computing systems as it spends heavily on artificial intelligence. A lease with Anthropic would move that idea from a possible future option toward a commercial test.

The company expects capital expenditures of $125 billion to $145 billion in 2026 because of AI infrastructure, equipment costs and additional data center capacity. Leasing part of that network could create revenue without forcing Meta to abandon its internal model-development plans. The spending remains unusually high.

Meta Cloud Strategy
Mark Zuckerberg told shareholders in May that outside companies approach Meta “almost every week” seeking access to its models or computing capacity. He said a cloud business was “definitely on the table,” but Meta has not announced one.
His remarks suggest Meta views outside demand as a safeguard if its infrastructure eventually exceeds internal requirements.

A contract would place two AI competitors in a commercial relationship, with Meta supplying the computing power required to train and operate Anthropic’s Claude models.

Anthropic needs advanced processors, reliable power and dedicated facilities as demand grows among subscribers and enterprise customers. Investors would also gain an early measure of Meta’s ability to recover part of its infrastructure costs.

The proposal follows Anthropic’s May agreement with SpaceX, led by Elon Musk, for access to the Colossus 1 data center and more than 220,000 Nvidia processors.
Reports valued that contract at nearly $45 billion over three years. The earlier contract was much larger.

Anthropic raised usage limits for paying customers after announcing the SpaceX capacity deal. That agreement showed how scarce computing resources had become, and why AI developers increasingly rent infrastructure from companies they otherwise compete against.

Stablecoins are emerging as the primary tool for facilitating micropayments among AI agents, according to recent findings by Visa and blockchain research firm Artemis. As artificial intelligence continues to intertwine with digital commerce, stablecoins, with their low transaction costs, are pinpointed as a viable method for frequent, machine-driven financial exchanges. Although card networks are expected to dominate consumer payments, stablecoins offer a complementary solution for automation-focused transactions.

How are stablecoins reshaping digital micropayments?
The joint research delves into the impact of AI agents—automated systems capable of conducting transactions—on digital payments. The study differentiates between macro-commerce, like hotel reservations managed by consumers, and micro-commerce, characterized by frequent, low-value transactions handled solely by software.

Stablecoins, pegged to fiat currencies like the USD, are especially adaptive to the micro-commerce realm. Their blockchain roots ensure seamless, low-cost transactions, making them well-suited for the high-frequency, small-scale exchanges that AI systems generate.

These micro-exchanges typically occur in the background as applications interact through APIs or exchange computational services. Given the inefficiency of traditional payment networks for these nominal transactions, stablecoins present a cost-effective solution safeguarding business-to-business transactions.

Visa and Artemis emphasized that stablecoins’ minimal transaction costs make them a compelling choice for recurring software-based payments, setting them apart from fixed-fee card networks.

Is dual-network the future of AI payments?
Visa envisions a future where payment systems integrate both conventional card networks and blockchain-backed stablecoins, fostering a dual-network approach. In such systems, AI agents intelligently select the payment method suited for each transaction: legacy card networks for consumer transactions, and stablecoins for automated machine payments.

Integration of stablecoin technology by established financial institutions highlights a trend where traditional payment firms increasingly mesh with the digital asset world. Visa, known for its processing prowess, has launched initiatives that connect standard transaction procedures with blockchain settlements. Concurrently, crypto-native companies invest in fortified security measures to align with traditional norms.

Consumer transactions prefer card networks due to their broad acceptance and dispute solutions.
AI agent micropayments benefit from stablecoins’ low costs and rapid settlements.
What are the regulatory challenges facing AI micropayments?
Regulatory ambiguity and handling disputes pose substantial barriers despite blockchain’s potential in AI micropayments. Current regulations emphasize human oversight, which conflicts with transactions that are entirely software-managed.

Chargebacks and consumer protections, designed for larger transactions, are ill-equipped for the continuous, automated exchanges in AI-driven commerce. Successful adoption requires payment systems to devise solutions for managing risks and disputes unique to machine-generated transactions.

Visa has taken steps to boost AI-driven and blockchain-based payment methods, joining forces with industry leaders like Mastercard and Coinbase. Through this collaboration, Visa supports open stablecoin protocols, reflecting its drive to enhance digital asset payments globally, primarily in automation and micro-transactions.

Stablecoins are broadening their reach with new partnerships and card programs compatible with blockchain settlements, reinforcing predictions that these digital currencies will underpin future AI-driven commerce.

The Ethereum Foundation is navigating a pivotal restructuring phase in response to ongoing scrutiny over its strategic priorities and leadership. In 2026, the foundation faced considerable criticism regarding its focus and capability to enhance the core Ethereum network, particularly against the backdrop of evolving layer-2 technologies.

Leadership Shifts and Staff Exodus?
These criticisms culminated in a series of substantial leadership changes. Early in the year, industry observers noted the exit of nine senior officials, prompting questions about the foundation’s future trajectory. Despite concerns, the foundation asserted that the turnover was part of a broader strategy to refine organizational efficiency, rather than signaling distress.

What Prompted the June Revamp?
June marked a significant milestone in the foundation’s evolution as co-executive director Hsiao-Wei Wang’s resignation precipitated an extensive overhaul. This restructuring effort saw the workforce shrink by 20%, with 54 roles being dissolved to cut operational costs by 40%. The remaining personnel were reorganized into five core groups focusing on mission-critical projects, reflecting a strategic pivot aimed at streamlining resource allocation within the ecosystem.

Following the restructuring, the Ethereum landscape witnessed the emergence of independent entities to assume roles formerly managed by the foundation. One such organization, ETHLabs, was launched to support protocol research and innovative product development. Similarly, Ethereum Institutional and EthSystems were formed to cater to enterprise integration and privacy-focused infrastructure development, respectively.

Workforce reduced to approximately 216 employees post-restructuring.
Annual budget expenditures cut down to 60% for improved sustainability.
Consolidation into five core operating groups to reinforce mission alignment.
The organizational evolution of the Ethereum Foundation points to a decentralized approach, where various specialized groups now assume greater responsibility for innovation.

“The introduction of groups like ETHLabs and Ethereum Institutional has shifted technical research, enterprise support, and infrastructure development into dedicated organizations beyond the foundation, suggesting a new era of decentralized management for the Ethereum ecosystem.”