Today on The Systematic Desk, following recent industry warnings about AI governance gaps, we're tracking the rapid institutional build-out of AI agent infrastructure. Major tech players are shipping foundational governance layers, while Morgan Stanley becomes the first Wall Street bank to open its core wealth management platforms to external AI agents.
At its Data + AI Summit, Databricks unveiled a comprehensive developer agent platform and open-sourced Omnigent, a 'meta-harness' for orchestrating and governing multiple AI coding agents. The expanded 'Agent Bricks' platform provides infrastructure to manage model choice, context, and control, addressing critical enterprise challenges like cost escalation, security, and the 'hidden technical debt' of deploying agents at scale.
Why it matters
This move from Databricks is significant because it shifts the enterprise AI focus from building individual agents to governing swarms of them. For building and managing trading infrastructure, this provides a blueprint for a control plane that can manage costs, enforce policies, and sandbox different research and execution agents securely. The open-sourcing of Omnigent offers a practical framework for building a robust, governable AI development environment.
Morgan Stanley announced Tuesday it will become the first major Wall Street bank to allow external AI agents to access data from its ShareWorks and Equity Edge stock administration platforms. The integration utilizes the Model Context Protocol (MCP) as an open-source standard to provide AI agents with secure, context-aware access, aiming to scale services for its 3,400 corporate clients.
Why it matters
This is a landmark moment for AI in institutional finance, demonstrating a move from internal tools to allowing trusted external agents access to core systems. The adoption of the open-source MCP is particularly notable, suggesting an industry-wide convergence on a standard for interoperability. For builders of financial infrastructure, this signals that future systems will need to be 'agent-native,' with secure, auditable APIs designed for programmatic access.
Announced Tuesday at the Data + AI Summit, Salesforce and Databricks are expanding their partnership to create a shared, governed foundation for both human and AI agent workflows. The collaboration focuses on unifying data and business context securely, featuring new governance for Zero Copy infrastructure, Federated Search, and integrations with the Model Context Protocol (MCP) to ensure AI agents operate with trusted, compliant data.
Why it matters
This partnership tackles a core problem in enterprise AI: securely connecting siloed data with the business context needed for agents to perform useful work. By building a unified governance layer that spans both platforms, they are creating the plumbing required to deploy compliant AI agents at scale. For developing any kind of financial system, this model of federated, governable data access is a critical architectural pattern.
Amazon Web Services announced Tuesday the integration of AI coding assistants like Kiro and Cursor into its Amazon SageMaker Unified Studio. The integration provides automatic AI steering and support for the Model Context Protocol (MCP) server, allowing AI agents to understand project context and data structures within governed cloud environments from the first prompt.
Why it matters
This brings powerful AI-assisted development directly into the secure, governed environments where institutional data lives, removing a major friction point for adoption. The native support for MCP signals its growing role as a standard for providing AI agents with secure, context-aware access to proprietary data. For quantitative research workflows, this means quants can use state-of-the-art AI assistants on sensitive data without complex workarounds.
Bybit launched options trading for Tether Gold (XAUT) on Tuesday, becoming the first crypto exchange to offer options on a tokenized physical asset. Developed with Orbit Markets, the offering includes a Request for Quote (RFQ) system designed for institutional clients, allowing for customized options contracts on gold settled on-chain.
Why it matters
This development bridges the gap between traditional gold derivatives and the crypto market structure. For systematic traders, it unlocks new, capital-efficient strategies for expressing views on gold's volatility and direction using a tokenized underlying. The institutional RFQ system is a key piece of infrastructure, suggesting this is not just a retail product but a serious attempt to build a liquid, professional market for on-chain gold derivatives.
Building on the CFTC's approval of the first U.S. onshore perpetuals on KalshiEX we tracked recently, Kraken has launched its own CFTC-regulated perpetual futures trading for eligible U.S. customers. Effective June 15 and enabled by its acquisition of Bitnomial and NinjaTrader, the move provides a compliant venue for leveraged trading on major cryptocurrencies.
Why it matters
The availability of regulated, onshore perpetual futures fundamentally changes the market structure for U.S.-based systematic funds. Following KalshiEX's breakthrough, Kraken's entry provides a major compliant alternative to offshore exchanges, potentially redirecting significant liquidity and altering counterparty risk models.
South Korean financial group Mirae Asset Global Investments ($721B AUM) signed an MOU with Ondo Finance on Tuesday to tokenize its Global X ETF lineup. The partnership will begin with a tokenized share class of a Hong Kong-listed covered-call ETF in Q3 2026, marking the first time a major Asian asset manager will tokenize its own listed funds for on-chain distribution.
Why it matters
This is a significant step forward for tokenization. Unlike many early RWA projects that were wrappers created by third parties, this involves a major asset manager directly creating native tokenized share classes of its existing, successful ETFs. It treats the blockchain as a legitimate distribution and settlement channel, providing a powerful case study for how traditional fund structures can be integrated with on-chain infrastructure.
In stark contrast to the recent failure of the xStocks tokenized SpaceX offering to secure underlying equity, Coinbase confirmed plans Tuesday to launch strictly 1:1 backed tokenized U.S. stocks for non-U.S. markets. Rolling out next month, the offering features on-chain dividend payments and an explicit 'no derivatives, no IOUs' structure, aiming to unify liquidity across its spot, derivatives, and institutional platforms like Deribit.
Why it matters
Coinbase's entry with a fully backed, dividend-paying model sets a rigorous compliance standard in the immediate wake of the xStocks settlement collapse. For offshore fund structures, this creates a major new source of regulated, on-chain U.S. equity exposure. The plan to pay dividends on-chain automates a traditionally manual part of asset servicing, providing a template for other tokenized assets.
Following the SEC's proposal on June 11 to rescind Rule 611 of Regulation NMS (the 'trade-through' rule), new analysis highlights the structural implications for tokenized equities. The move is seen as a deliberate step to clear a path for on-chain automated market makers (AMMs) and other DeFi trading models, which cannot easily comply with the rule's routing requirements.
Why it matters
This is a recurring thread, but the new analysis frames the SEC's move not just as deregulation but as a conscious architectural choice to enable a specific type of market structure. Rescinding Rule 611 would remove a key obstacle preventing tokenized equities from being traded natively in DeFi, directly impacting the design and viability of on-chain trading venues for tokenized funds.
The SEC on June 12 approved a proposal for the T. Rowe Price Active Crypto ETF (TKNZ) to be listed on NYSE Arca. The fund represents a significant step for regulated crypto products, as it will be actively managed and can hold a portfolio of 5 to 15 different digital assets, including BTC, ETH, SOL, and XRP, moving beyond single-asset spot ETFs.
Why it matters
This approval signals the SEC's growing comfort with more complex, actively managed crypto products from established financial players. It provides a regulated, diversified on-ramp for institutional investors like pension funds, which may have mandates that prevent them from holding single assets directly. The approval of a multi-asset fund also legitimizes a broader swath of the crypto market beyond just Bitcoin and Ethereum.
Biaurum Capital Fund SP has launched a new Cayman-regulated global macro fund, which will invest in precious metals, digital assets, and derivatives. The fund is operating on the CV5 Capital platform, which provides institutional-grade infrastructure for emerging managers, and is targeting a minimum investment of US$100,000.
Why it matters
This launch serves as a current case study in emerging manager operations. Using a platform like CV5 Capital highlights the build-vs-buy tradeoff for fund infrastructure, allowing managers to offload operational complexities. The fund's mandate, which combines traditional macro assets with digital assets within a regulated Cayman structure, reflects an increasingly common strategy for new launches.
The government of the Bahamas is moving forward with its national digital identity system, which will serve as a foundation for streamlining government services and improving identity verification. The initiative is paired with forthcoming AI legislation and a national digital skills program aimed at boosting the country's digital economy.
Why it matters
While not directly about fund structures, this development is a key indicator of the Bahamas' commitment to building modern, digital-native public infrastructure. For operators considering the jurisdiction, a functioning national digital ID system significantly improves the ease of doing business, from company formation to personal banking and residency applications, impacting the practical quality-of-life and operational friction of being based there.
AI Agent Infrastructure Becomes a Product Category Major tech platforms including Databricks, AWS, and Salesforce are rolling out comprehensive governance and orchestration layers for AI agents, moving beyond model development to offer tools for context management, security sandboxing, and cost controls. This signals the maturation of agentic AI from experimentation to enterprise-grade production.
Institutional Finance Adopts Open Protocols for AI Agents Morgan Stanley is the first major bank to open its core stock administration platforms to external AI agents, notably using the open-source Model Context Protocol (MCP) as the integration layer. This follows similar moves by Robinhood and is now being adopted by AWS, pointing toward MCP becoming a standard for secure, context-aware data access in finance.
Tokenized Gold Derivatives Bridge TradFi and DeFi Bybit is launching options trading for tokenized gold (XAUT), including an RFQ system for institutional clients. This creates a new derivatives market on a physical, tokenized asset, expanding the toolkit for systematic strategies beyond pure crypto assets.
The Onshoring of Crypto Derivatives Accelerates Kraken's launch of CFTC-regulated perpetual futures for U.S. traders marks a significant milestone in bringing crypto market structure onshore. This follows the approval of the first onshore perpetuals and signals a shift that will impact liquidity, market making, and infrastructure choices for U.S.-based funds.
Tokenized Fund Distribution Expands to Asia South Korean financial giant Mirae Asset is partnering with Ondo Finance to tokenize its Global X ETFs. This is the first instance of a major Asian asset manager tokenizing its own listed funds, signaling a strategic move to use blockchain as a new distribution channel for mainstream investment products.
What to Expect
2026-06-29—Launch of the Security Token Offering (STO) Foundation, a global membership organization for the RWA tokenization industry.
2026-07-01—MiCA's transitional window for crypto licensing in the EU ends, potentially forcing unlicensed operators to exit.
2026-07-01—Coinbase plans to roll out tokenized stocks for non-U.S. users.
July 2026—DTCC is targeting a limited production pilot for tokenizing real securities.
How We Built This Briefing
Every story, researched.
Every story verified across multiple sources before publication.
🔍
Scanned
Across multiple search engines and news databases
376
📖
Read in full
Every article opened, read, and evaluated
168
⭐
Published today
Ranked by importance and verified across sources
12
— The Systematic Desk
🎙 Listen as a podcast
Subscribe in your favorite podcast app to get each new briefing delivered automatically as audio.
Apple Podcasts
Library tab → ••• menu → Follow a Show by URL → paste