News feed

Framework Ventures has launched a new $400 million fund focused on financing capital-intensive industries such as artificial intelligence (AI), robotics, and energy through blockchain technology. Co-founder Michael Anderson stated that the firm is shifting its focus from serving cryptocurrency users to addressing real-world financing challenges, leveraging tools like tokenization and stablecoins.

Anderson believes that the biggest opportunities in blockchain are emerging as the technology evolves beyond its initial speculative uses. Instead, tokenization can provide better financing options for assets like GPUs and other computing hardware, which have traditionally struggled to find adequate investment vehicles. "We have the capital on-chain to finance this industry," said Anderson.

The fund's approach includes funding projects that utilize blockchain as financial infrastructure to solve real-world problems. An example includes the firm’s investment in Daylight, a company financing residential solar projects, and Uranium Digital, which is creating a tokenized marketplace for uranium. This shift also reflects a broader trend in the digital asset sector, where traditional financial institutions are beginning to incorporate blockchain technologies into their operations.

Anderson noted a significant change in the profile of founders launching crypto-related ventures, stating that many now come from backgrounds in traditional finance, energy, or industrial technology, offering expertise that has been lacking in previous cycles. This trend underscores a potential move towards sustainable business models centered around technological utility rather than mere speculation.

As the digital asset landscape evolves, Anderson poses a thought-provoking question: "What if 2021 was the aberration and we're now moving toward fundamental utility?" With a focus on providing practical solutions through tokenization and on-chain capital, Framework Ventures positions itself at the forefront of this transitional phase in the crypto and blockchain space.

Jansen Teng, CEO of Virtuals, recently outlined the company's vision for AI agents as evolving into autonomous economic actors. Initially focused on creating AI for gaming, Virtuals has broadened its scope to build infrastructure for what it calls an “agent society.” This includes developing digital and physical agents, supporting agent coordination, and establishing capital formation and governance systems.

Teng emphasizes that the new phase of AI agents may shift focus from mere conversational capabilities to active participation in economic transactions. He believes that as these agents gain the ability to manage wallets and engage in trading, they will operate more independently, engaging in activities such as hiring other agents or coordinating economic efforts.

"The long-term vision is a parallel society where agents can participate in a permissionless economy," Teng stated. This ecosystem could allow agents to perform numerous tasks without the need for constant human intervention. However, it also introduces significant risks, such as incorrect user intent, service fulfillment failures, and potential scams. To mitigate these risks, Virtuals is exploring various solutions, including intent verification systems, escrow-based transaction standards, and reputation frameworks to build trust within agent networks.

By creating systems that allow agents to manage their economic activities, Virtuals aims to establish a decentralized alternative to traditional financial ecosystems that are currently under development by financial institutions. Teng compares this approach to the foundational shifts brought about by Bitcoin and Ethereum, proposing that fully on-chain agents could operate without needing to navigate identity and regulatory hurdles until they interact with conventional financial infrastructures.

Looking ahead, Teng anticipates the merging of robotic technologies with digital agent economies. Virtuals collaborates with various stakeholders, including robotics startups and educational institutions, to explore real-world applications for humanoid robots that could work alongside digital agents. This multidisciplinary approach targets software solutions, commercialization, and data collection, rather than hardware production, suggesting a future where AI manages operational functions in businesses while physical robots handle customer-facing roles.

Palo Alto-based blockchain startup Story Protocol has rebranded itself as DATA Foundation, focusing on the challenges surrounding artificial intelligence (AI) copyright violations. The company raised $140 million, primarily from Andreessen Horowitz's a16z crypto, to develop infrastructure that ensures the legitimacy of datasets used to train AI models.

The new DATA Network intends to deliver a clear documentation trail through its Trace platform, which offers cryptographic receipts detailing dataset provenance, licensing, consent, and payment history. This approach seeks to address the growing scrutiny that AI developers and large technology firms face over dataset ownership and the necessity for proper consent in data collection.

Avi Patel, founder of Kled AI, joins DATA Foundation as chief data officer, further emphasizing the company's commitment to the AI data sector. Patel mentions that the integration with Kled AI's marketplace allows for the registration of 1.1 billion user-contributed records, creating a comprehensive database for AI to draw from.

The rebranding follows previous criticisms and operational delays the startup faced, particularly regarding token unlock events earlier in the year. Patel highlighted that the firm’s restructuring aims to build a specialized focus on AI data legality rather than a broader approach to intellectual property. He notes that the industry suffers from legal challenges stemming from data scraping without proper consent, making it imperative for AI developers to have confirmation of dataset legitimacy.

"Labs won't license data they can't verify," said Patel, highlighting the necessity for fraud-detection protocols that confirm data authenticity, originality, and compliance to prevent the use of pirated or AI-generated datasets.

The Trace platform is designed to create immutable records of data contributions, ensuring that what is made public are not the datasets themselves but rather receipts containing essential information such as content hashes, consent terms, and timestamps. This feature protects the underlying data while providing a verifiable history that can be audited.

To further sustain its mission, DATA Foundation is simultaneously developing an application called Numo, which rewards contributors for submitting validated data with stablecoin payments. This allows for real-time compensation tied to the completion of transactions, enhancing the financial incentives for data providers.

As DATA Foundation moves forward, it faces the challenge of establishing itself amidst increasing competition in the AI data compliance space while aiming to resolve the copyright issues that bedevil many operating in this field.

On June 25, 2026, BlackBerry experienced a stock surge of nearly 23% thanks to a strong earnings report that exceeded expectations and projected higher future growth. The company is shifting its focus away from its historical role as a cellphone manufacturer, instead positioning itself as a provider of essential software for the "Physical AI" and robotics sectors.

The catalyst for this transition is BlackBerry's QNX operating system, recognized for its reliability and safety. This software provides the framework used by significant chip manufacturers like Nvidia and AMD to construct intelligent systems, including smart vehicles and autonomous robots. CEO John Giamatteo noted during an earnings call that, "As intelligent machines become increasingly autonomous and operate around people, the requirements for safety, security, reliability and real-time determinism become even more important." By emphasizing the deterministic nature of QNX, which differs from more common probabilistic AI systems, BlackBerry aims to assert its niche in scenarios where system failures are intolerable.

The firm's transformation from a once-dominant mobile device company to a critical software supplier reflects broader trends in technology, as industries increasingly rely on robust and secure solutions for AI-driven operations. Notably, the company has retained elements of its legacy in cybersecurity, applying the cryptographic principles that secured its mobile devices to the software solutions it offers today. This shift highlights BlackBerry's continued relevance as it adapts to the evolving landscape of technology.

Investors and analysts are taking note of this shift towards AI, which is rapidly gaining traction across various sectors. The market response indicates a favorable outlook as BlackBerry's advancements in software align with the growing demand for security and reliability in autonomous systems. With the surge in stock value following the recent report, the company appears well-positioned to capitalize on the AI boom.

Looking ahead, stakeholders may want to watch how BlackBerry continues to innovate within the AI context and whether its historical reputation for security can translate to sustained operational success in this new arena.

Cerebras Systems, an AI chipmaker that completed a significant IPO in May, has reported its first earnings results, which include a 92% year-over-year revenue increase to $193.4 million. However, following the release, the stock dropped by 11% in after-hours trading as investors reacted to guidance suggesting a decline in core gross margin for the next quarter.

In its forward-looking statement, Cerebras forecasted second-quarter revenue of approximately $194 million but expects core gross margins to decrease to between 36% and 38%, compared to 46.5% in the first quarter. This deterioration in profit margin appears to have overshadowed the revenue growth and has led to a notable sell-off in shares after the earnings announcement.

"Despite a significant revenue jump and a narrower net loss than anticipated, the lower gross margin guidance has raised concerns among investors," said a market analyst. This statement highlights the weight investors place on profit margins as indicators of future performance.

Cerebras, which raised $6 billion during its IPO at a price of $185 per share, saw its stock surge to a peak of $385 shortly after the market debut. As of the latest trading, shares are priced at $201.55.

AI-driven security tools are becoming increasingly accessible and could significantly affect the cost and efficiency of smart contract audits within the crypto industry. A notable example is Mythos, an AI system that autonomously identifies vulnerabilities in code, potentially reducing the cost of audits to near zero, as stated by Alexander Urbelis, chief information security officer at ENS Labs, who noted, "It pushes the price of a basic audit toward zero." This advancement allows for rapid assessments that could previously only be performed by skilled professionals and at significant expense.

Traditionally, smart contract security has faced financial barriers, limiting extensive audits. AI technology like Mythos marks a shift in the auditing approach, as these systems possess reasoning capabilities—enabled, for instance, by analyzing the intended function of the code versus its actual performance. This newfound capability is especially relevant in the crypto landscape, where smart contract transparency paired with substantial bug bounty budgets can widen the net of vulnerability detection.

David Schwed, COO of SVRN and founder of a cybersecurity master's program at Yeshiva University, highlighted the transformative potential of these AI models: "These models now operate the way a human attacker does." The implication is that AI tools are capable of continuous security monitoring, allowing for ongoing updates and suggested fixes rather than relying on a singular, costly audit process. Schwed emphasized that if security evaluations translate into affordable and continuous checks, it could redefine expectations within the industry regarding due diligence and oversight.

As these AI security solutions become ubiquitous, the industry may face new standards for audits. Urbelis pointed out that the availability of advanced AI analysis might shift liability, stating, "A clean AI report will be seen as no defense." Investors could soon expect AI-assisted audits before funding projects, suggesting a potential shift in how negligence is perceived in crypto security decisions.

However, experts caution that while AI tools can assist in identifying technical flaws, they cannot fully replace human oversight. Many significant security breaches have resulted from social engineering and operational failures rather than just coding errors. For instance, Urbelis referenced a recent breach involving the Drift platform, emphasizing that the human element in security is crucial: "The smart contract did exactly what it was told; the authority behind the instruction was what was compromised and abused."

Consequently, while AI has the potential to alleviate some security challenges by making vulnerability detection faster and more cost-effective, it does not eliminate the need for experienced human auditors to assess complex economic and incentive structures that can lead to breaches. Schwed reiterated this, remarking that simply relying on AI tools without proper human knowledge would not guarantee safety: "If the person running the tool can't evaluate what comes back, you haven't bought security, you've bought a false sense of it."

Blockchain firm Alchemy has introduced its AgentCard, which integrates with Visa's payments system, allowing AI agents to make online transactions on behalf of consumers. Each AgentCard functions as a virtual identity and payment card tailored for AI agents, enabling them to use Visa-issued tokens while maintaining existing rewards, credit lines, and benefits.

The integration with Visa Intelligent Commerce enables AgentCard holders to book vacations, order groceries, or renew subscriptions automatically, without direct human involvement in the checkout process. Each AI agent provisioned through AgentCard receives a unique email address and phone number, ensuring a complete identity layer for service access and verifications.

The concept of "agentic commerce" represents a shift in how consumers might engage in online shopping by delegating tasks to AI. "Every major computing shift has produced a new kind of economic actor," said Nikil Viswanathan, co-founder and CEO of Alchemy. He adds that AI agents are the next logical step in this evolution, stating, "AgentCard is how that starts." This move aligns Alchemy with other major firms like Stripe and Mastercard, which are also exploring similar spaces.

With this newly established connection to Visa, AgentCard aims to become a significant player in the nascent field of agent-native payment protocols. The technology behind AgentCard conducts transactions via an optimized routing layer, identifying the best payment methods for each situation and using single-use tokens in instances where alternative protocols are not supported.

The full impact of this service rollout remains to be seen, particularly as adoption gathers pace among users and businesses alike. Observers will be watching how effectively AgentCard can streamline the consumer experience and how it influences the broader landscape of digital payments.

HIVE Digital Technologies (HIVE) has announced a $220 million, three-year contract to provide AI computing infrastructure in Canada. This deal involves deploying 2,304 Nvidia Grace Blackwell Graphics Processing Units (GPUs) at Bell Canada’s AI Fabric facility in Merritt, British Columbia, which will support enterprise AI models for various government and corporate clients.

This new focus on AI marks a significant shift for HIVE, which has historically concentrated on bitcoin mining. The company expects this contract to contribute approximately $70 million in annual recurring revenue, which will boost HIVE's total contracted high-performance computing (HPC) revenue beyond $100 million.

The AI infrastructure will remain on Canadian soil, aligning with the Canadian government’s efforts to diminish reliance on foreign-controlled AI technology. "This deployment solidifies our commitment to building a fully sustainable computing environment in Canada," said HIVE’s executive team.

The project is slated to launch between late 2026 and early 2027, highlighting HIVE's potential for substantial growth as it ventures deeper into the AI sector while pulling back from traditional cryptocurrency mining activities.

Investors may want to watch HIVE's upcoming activities closely, as the success of this infrastructure deal could indicate further strategic pivots in the broader tech landscape and HIVE's role within it.

Stables Introduces AI-Native Payment Middleware

Stables, a startup based in Singapore, aims to revolutionize cross-border trade in Asia by developing a specialized payment middleware designed for automated software systems. The company focuses on facilitating stablecoin transactions across the region, leveraging its new technology to overcome the limitations of existing financial infrastructure.

The Rise of Machine-to-Machine Payments

CEO Bernardo Bilotta predicts a shift in the landscape of commerce, emphasizing that artificial intelligence (AI) systems will increasingly handle financial transactions. According to Bilotta, “Between now and the next five years, I think the entirety of commerce will be moving through AI agents.” This reflects a broader transformation, as automated agents potentially take precedence over human users in the financial ecosystem.

Addressing Infrastructure Gaps

Currently, 60% of stablecoin payments are made through a fragmented payment structure in Asia, which presents challenges for compliance and efficiency. Bilotta comments that legacy banking protocols were not designed to accommodate autonomous AI systems, which struggle to navigate standard compliance checks without human intervention. To address this gap, Stables is embedding an Anthropic Model Context Protocol (MCP) server into the payment infrastructure to streamline compliance, foreign exchange, and settlement processes.

The Trade Landscape and Future Projections

Globally, stablecoins transacted approximately $35 trillion last year, with projections indicating this figure could soar to $700 trillion by 2035. This rapid growth underscores a pivotal shift in global trade dynamics. The Asia-Pacific B2B e-commerce market alone is expected to exceed $28.9 trillion in market value by the end of the year, growing at a rate of 15% annually, according to business data from the U.S. International Trade Administration.

Conclusion: The Future of Stablecoins

Stables argues that the next phase of stablecoin adoption will be driven more by businesses and fintech companies than by retail traders. Bilotta states, “The next phase of stablecoin adoption won’t be driven by crypto traders. It will be driven by businesses, fintechs, and eventually AI systems moving real money in the real economy.” As the infrastructure evolves, the role of AI in facilitating seamless transactions could redefine financial interactions across sectors.

Bittensor has submitted a proposal called Root Reborn aimed at restructuring how its TAO validators function. The proposed changes would allow validators to select specific subnets to support instead of automatically selling subnet tokens for TAO. This new model intends for validators to reinvest rewards into chosen subnets, fostering capital growth and potentially preventing downward pressure on token prices.

The Root Reborn proposal represents a significant departure from current practices. Presently, to provide yield to TAO stakers, the network sells the rewards owed and swaps out subnet tokens for TAO at each block. However, this model has the drawback of consistently selling valuable subnet tokens, thereby negatively impacting their market prices.

Under the new plan, validators would act similar to fund managers, choosing which subnets to back. The yield that would otherwise be sold could instead be invested back into chosen subnets, allowing for capital compounding over time. Stakers still receive their yields while retaining the option to cash out into TAO whenever they choose.

This approach not only turns validators into active curators of capital but also incentivizes them to choose wisely, as their decisions could influence the influx of funds into those subnets. Validators will have to evaluate which subnets to back based on performance, potentially encouraging more engagement and thoughtful investment strategies.

As for the technical side, the proposal has been submitted for a test network on GitHub and has undergone an initial automated review that flagged and subsequently addressed two significant issues before its mainnet release. One of these issues involved handling large data amounts, and the other pertained to a payout pathway that might underpay stakers if a subnet were to shut down.

In recent data, TAO's token has seen a decline, dropping 28% over the past 12 months, in contrast to Bitcoin's 38% decrease in the same timeframe. Currently, staking yields for TAO are around 17%, indicating some level of return for stakers, though this may attract scrutiny as the network adjusts its policies and practices.

The Root Reborn initiative signifies an important shift in how decentralized networks can structure their economic models to sustain long-term viability and encourage positive price action for subnet tokens.

John Hoffman, Managing Director and Head of Product Portfolios at Ondo Finance, explains how tokenization is positioning itself as a transformative force in financial markets. He parallels current advancements in tokenization with the inception of exchange-traded funds (ETFs), emphasizing that the market for tokenized assets has outpaced traditional estimates, expanding to over $33 billion.

Hoffman predicts that the total value of tokenized assets could reach trillions in the coming decade, driven by the increasing integration of artificial intelligence (AI) into investment strategies. "The future of markets are onchain," said Hoffman. He advocates for a future where AI agents autonomously transact and allocate capital through tokenized financial products.

In his view, tokenization represents the digitization of capital markets. Hoffman recalls how ETFs, initially met with skepticism, have transformed into a nearly $20 trillion asset class. He believes the tokenization movement is evolving even more rapidly, suggesting that as markets digitize, they grow larger.

Hoffman elaborates that for AI-driven financial solutions to thrive, there is a need for tokenized assets alongside robust blockchain infrastructure and management strategies capable of operating onchain. Ondo Finance is advancing in this direction, currently offering tokenized U.S. Treasury products and planning to expand its offerings to stocks, ETFs, and perpetual futures.

Hoffman envisions a landscape where financial portfolios are continuously managed and adjusted in real-time, adapting to market changes and data fluctuations. "Our end state will be portfolios that are professionally managed, real-time and adjusting to market circumstances and data changes," he said, highlighting Ondo’s goal to provide a trusted platform for onchain investment portfolios.

Ripple launched the XRPL AI Starter Kit, a collection of tools enabling developers to create AI agents that can make payments via the XRP Ledger using both XRP and Ripple's stablecoin, RLUSD. This initiative targets the expanding x402 machine-to-machine payment framework, which currently favors USDC, having processed over 120 million transactions across 14 different blockchains.

The XRPL AI Starter Kit provides essential features such as access to XRPL documentation through a managed server, functionalities for wallet creation and balance checks, and support for x402 payments. Ripple aims to attract developers by emphasizing the XRP Ledger's notable advantages, such as first-rate speed of transaction settlement and low transaction costs.

Ripple's motivation for entering this market stems from the rapidly growing demand for reliability and efficiency in automated transactions. AI agents often require seamless payment systems to settle invoices, acquire API access, or transfer value without human intervention. The XRP Ledger claims to enable settlements within three to five seconds and features predictable fees, which can be significant for high transaction volumes.

The x402 protocol, initially introduced by Coinbase and now managed by the x402 Foundation under the Linux Foundation, is designed to facilitate machine payments through standard web requests. This system allows an AI agent to request a service, receive a payment request, make an on-chain payment, and then resubmit the request with proof of payment. By designing payments to function more like API calls, the infrastructure is purportedly easier for developers to integrate.

Despite Ripple's optimistic claims, the x402 market has recently been characterized by limited activity primarily involving stablecoins. A Chainalysis report indicated that x402 activity on Base surged from negligible levels in mid-2025 to over 100 million total transactions as of early June 2026. However, a substantial portion of this surge has been attributed to speculative activities surrounding PING, a meme coin experiment that utilized x402 payment capabilities.

As Ripple steps into this landscape, it faces inherent risks associated with the x402 protocol. Some academics have raised concerns about potential failure points in synchronizing payments across web and blockchain platforms. For instance, issues could arise in validating payments, mismatches in payment requests, or reusing outdated payment proofs, which could undermine trust in the system.

Despite launching new tools aimed at streamlining financial transactions for AI agents, Ripple has not disclosed specifics regarding customer adoption, transaction volumes, or any successful implementations at scale. As such, while the initiative holds promise, the actual efficacy and recognition of XRP and RLUSD in real-world applications remain to be clarified.

On June 13, 2026, the U.S. government ordered Anthropic to cease access to its advanced AI models, Fable 5 and Mythos 5, due to national security concerns linked to a potential jailbreak. This action led to immediate negative market reactions, with the price of Anthropic's perpetual contract on Hyperliquid falling approximately 3.7%, retreating to about $1,627 after reaching nearly $1,800 shortly after the models' launch.

The government's directive, received at 5:21 p.m. ET on June 10, required Anthropic to shut down access for both domestic and foreign users, following claims that it learned of a method to bypass the algorithms. Anthropic contested the government's actions, pointing out that the evidence presented is vague and based on a narrow vulnerability that other models, including OpenAI’s GPT-5.5, also identify without any bypassing method.

In its defense, Anthropic remarked, "The government has so far provided only verbal evidence of a potential narrow jailbreak" and that this standard could hinder new AI model deployments across the industry. The company expressed concerns that by taking this measure, the government risks stifling innovation in AI technology.

The market's response to the shutdown indicates growing concerns over Anthropic's pre-IPO prospects. With the perpetual contract's drop and current open interest at around $8.6 million—modest for an upcoming IPO—the prospects for a deal hinge on how the government responds to Anthropic's forthcoming technical rebuttal against the directive. As of now, the first significant question for investors and stakeholders remains whether the shutdown order will be reversed or modified.

What Happened

Coinbase has officially launched "Coinbase for Agents," a platform that empowers artificial intelligence assistants to connect directly to users’ Coinbase accounts for managing cryptocurrency trading and transactions. This new service, which became available on June 11, 2026, enables AI agents, including tools like ChatGPT and Anthropic's Claude, to execute trades, access market data, and eventually facilitate payments autonomously, all within user-defined parameters.

Why It Matters

The platform is part of Coinbase's vision for what it terms "agentic commerce," where AI could handle various financial operations for users. By integrating with its x402 payments protocol, the company seeks to allow AI agents to make seamless transactions for services such as accessing premium market research without requiring ongoing subscriptions or manual intervention. Coinbase has touted that the advent of such autonomous agents could account for up to 20% of e-commerce activities by 2030.

To ensure security while utilizing AI agents, Coinbase has established several safeguards. Users can specify controls such as spending limits and the types of services the AI can access, thereby protecting against unauthorized transactions.

“AI agents will operate within isolated portfolios, which safeguards user assets and enhances control over automated transactions,” said Coinbase in their release of the new platform.

What to Watch

Looking ahead, it will be important to monitor user adoption of Coinbase for Agents and how effectively it manages security concerns. As Coinbase plans to expand the capabilities of AI agents to include equities and prediction markets, how these developments impact user trust and market dynamics will be significant. Moreover, the effectiveness of AI in trading strategies will also warrant observation, particularly as the project scales.

Tether Investments has taken the lead on a $1.4 billion funding round for Neura Robotics, a German firm specializing in artificial intelligence (AI), particularly in developing humanoid robots. This investment is described as one of the largest in the sector of physical AI to date and is expected to boost Neura’s valuation to between $9 billion and $12 billion.

Neura Robotics aims to produce 5 million AI-powered humanoid robots by 2030. The initiative includes equipping each robot with a Tether-linked digital wallet that would enable them to autonomously receive payments for their work and make electronic transactions. Neura's plans underscore a shift as AI technology transitions from theoretical applications into practical, physical use cases.

David Reger, founder and CEO of Neura Robotics, emphasized this transition, stating, "AI is moving from the digital world into the physical world." Tether's involvement marks a strategic move as it diversifies its investment portfolio beyond the cryptocurrency realm. Under the leadership of CEO Paolo Ardoino, Tether has begun to invest in various industries, including agriculture, brain tech, and sports, having reported over $10 billion in profits in the first nine months of 2025.

As Tether continues to expand its influence in technological sectors, its collaboration with Neura Robotics demonstrates a growing intersection between digital currencies and physical technologies. Investors and industry observers will be watching how this funding impacts the development of AI technologies and the integration of digital wallets in robotics.