LSEG plans new venue launch, Talos integrates with Kalshi, and more

After more than a month of football (or soccer to those in the US), the 23^rd edition of the FIFA World Cup has come to an end. A big congratulations to Spain! As most of the matches were in the wee hours of the morning for me—between 1 am and 5 am—I am thankful I now have my sleep back.

Announced this week

S&P launches service for AI agents and LLMs to access data

S&P Global has launched Adaptive Retrieval, a service allowing customer AI agents and large language models (LLMs) to access and assemble licensed S&P Global data using natural language queries. Through the S&P Global AI Data Portal, Adaptive Retrieval lets AI agents and LLMs pull data from many different sources at once and automatically handle requests that involve multiple datasets.

S&P said the service is suited for complex, multi-step tasks, including in-depth research and report generation.

LSEG to launch LSE 24

London Stock Exchange Group plans to launch London Stock Exchange 24 (LSE 24), a new 24/5 trading venue. LSE 24 is designed to support near-continuous trading from Monday to Friday from 5 pm to 7:50 am, with a 30-minute pause between 6:30 pm and 7 pm to apply end-of-day processes.

Trading will continue on the London Stock Exchange’s main market between 8 am and 4:30 pm. The venue will be built on LSEG’s financial market infrastructure. LSE 24 will be available for client testing by the end of 2026, with exchange-traded products launching as the first asset class in the first half of 2027, subject to regulatory approval.

HKEX adds Bursa Malaysia as recognized stock exchange

The Stock Exchange of Hong Kong Limited, a subsidiary of Hong Kong Exchanges and Clearing Limited (HKEX), has added Bursa Malaysia as a recognized stock exchange.

This will enable public companies with a primary listing on Bursa Malaysia’s main market to apply for a secondary listing in Hong Kong. Bursa Malaysia is now among four other Southeast Asia exchanges recognized by HKEX, including the Indonesia Stock Exchange, the Singapore Exchange, and the Stock Exchange of Thailand.

Nasdaq partners with Georgian central bank to modernize infrastructure

Nasdaq has partnered with the National Bank of Georgia, the country’s central bank, to modernize the treasury and financial markets infrastructure across Georgia’s banking sector. Five of the country’s largest commercial banks—Bank of Georgia, TBC Bank, Liberty Bank, Terabank, and Basisbank—will adopt the Nasdaq Calypso platform under a shared, common infrastructure model for the front-to-back trade lifecycle.

BMLL, Sigma AI partner for execution intelligence

BMLL, a provider of historical Level 3, 2, and 1 data and analytics for capital markets, has partnered with Sigma AI, a provider of real-time actionable intelligence for financial markets. The partnership will combine BMLL’s high-fidelity historical order book data with Sigma’s real-time engine to deliver a new intelligence layer that will allow market participants to instantly contextualize live market flow against deep historical benchmarks.

Broadridge integrates with Alpaca’s tokenization network

Broadridge Financial Solutions and Alpaca, an agent-first brokerage infrastructure, have integrated Broadridge’s governance infrastructure for retail and institutional investors into Alpaca’s Instant Tokenization Network.

Together, they enable traditional and tokenized equities to support the ownership rights, transparency, and operational integrity expected in traditional markets.

ExtractAlpha unveils transcripts AI model

ExtractAlpha has launched its Transcripts AI Model, an upgraded natural-language processing-based stock selection signal designed to extract predictive insights from earnings call transcripts.

The model uses contextual embeddings and machine learning to capture deeper language patterns associated with future stock performance. The signal is trained to predict forward equity returns and is designed for systematic investors seeking differentiated, low-turnover alpha from company communications.

It is available in English and Japanese. 

Allvue adds capabilities to OneVue private capital platform

Allvue Systems has launched Portfolio Intelligence and Deal Analytics on OneVue, its platform for private capital. The release delivers two capabilities: an AI-powered workspace for portfolio monitoring, and the first private credit benchmarks constructed from actual deal-level data.

Portfolio Intelligence gives credit teams a unified workspace for portfolio KPI monitoring, covenant headroom, peer benchmarking, and AI-assisted analysis—built directly on the data already maintained in the credit front office. Deal Analytics is powered by Allvue’s proprietary private credit dataset, spanning over 150,000 assets and securities and over 200 KPIs. It allows credit teams to compare their own deals against the private credit market on terms, covenants, and borrower financials.

Talos integrates with Kalshi

Talos, a provider of institutional digital asset infrastructure, technology and data, has announced an integration with Kalshi. Institutional clients will be able to trade Kalshi’s event contracts and crypto perpetuals using the same Talos interface used to trade digital assets.

Talos is launching two capabilities in line with the integration. For on-exchange trading, clients have access to the Talos algo trading suite. Talos’s multi-leg execution suite extends this to spread trading where clients can construct perp-to-perp and perp-to-spot spreads within a single order.

Also, the Talos request-for-quote platform will provide a block trading interface for trading large sizes off-exchange, connecting to Talos’s network of OTC liquidity providers.

What you might have missed from us

UK asset manager fixes data then brings in AI

The asset management industry is plagued with overlapping systems. Over time, getting one view of a firm’s data across all its systems gets more difficult. This was the case with Omba Advisory & Investments, which led it to seek a service provider to help consolidate its data and more.

Red alert: how Nasdaq’s Smarts became surveillance blind spot

On July 8, clients of Nasdaq’s trade surveillance software, Smarts, received notifications that the software failed to send alerts for some contracts due to incorrect multipliers embedded in the tool. The issue affected products on multiple venues, including BM&F Bovespa, Cboe, and Eurex.

One bank estimates that as many as 200 contracts were under‑monitored, potentially creating blind spots where market manipulation could have occurred.

Are recordkeeping rules ripe for updates?

In the latest Wrap, Rebecca Natale writes about the US Securities and Exchange Commission’s move to highlight recordkeeping in their rulebook. Broker-dealers should have an eye on SEC rules 17a-3 and 17a-4, which govern what records should be created and how long they must be maintained: three years or six years, depending on the type of communication.

IBM’s ‘multi-billion-dollar opportunity’ securing open-source software

IBM believes that the opportunity for open-source software security is in the multiple billions of dollars. Early adopters of Lightwell include banks like Bank of America, BNY, Citi, Goldman Sachs, JP Morgan, Morgan Stanley, Royal Bank of Canada, State Street, and Wells Fargo, among other firms.

During the firm’s second-quarter earnings call, IBM chairman, president and CEO Arvind Krishnan said IBM will measure itself by how many clients it can sign up over the next few months. “That is how aggressive we want to get, and we believe that that becomes a flywheel that multiplies,” he said.

Garage upstart turns Polymarket bets into benchmarks

Start-up Belief Systems is attempting to turn binary yes/no bets into rules-based benchmarks based on order book data from Polymarket. Founded by James Stewart, the firm has built 13 event-based indexes from Polymarket data that update every 30 minutes, covering themes from global conflict to AI acceleration.

During the global financial crisis, knowing your data was the edge. With AI, it’s the whole business

In this guest column, Peter Sanderson, CEO of SimCorp, says firms should understand their data and enterprise plumbing intimately in the age of AI.

In other news

New York Is Fighting the Wrong Energy Battle, The Wall Street Journal

In this letter responding to a Wall Street Journal article on New York Governor Kathy Hochul’s executive order pausing new AI datacenters for a year, Todd Pittinsky, a technology and society professor at Stony Brook University, called the move a “short-sighted response.”

He said that while large facilities can strain electricity, water and local infrastructure and protecting ordinary ratepayers is justified, suspending projects broadly is the wrong remedy. Developers should bear the costs of necessary upgrades rather than face an indiscriminate moratorium.

Also, a year-long freeze would merely drive investment, construction jobs and tax revenue to other states with more predictable timelines.

Given the energy needed to power and cool datacenters, I wonder whether we’ll see firms considering other, more energy-efficient locations rather than New York.