Top Companies' AI Moves

Today's Stories

1. 1

   BlackRock, Microsoft backers buy Aligned Data Centers for $40B

   BlackRock's Global Infrastructure Partners, UAE investment firm MGX, and the Artificial Intelligence Infrastructure Partnership (a group launched by BlackRock, Microsoft, and Nvidia in September 2024) have closed the acquisition of data center developer Aligned Data Centers for $40 billion（約6.4兆円）. The deal, first announced in October, represents AIP's first investment. The buyers will also provide $5 billion（約8000億円） in additional capital for expansion. Aligned owns over 6.4 gigawatts of operational and planned data center capacity across 51 locations, primarily in the U.S. with others in Mexico and South America. The move signals major institutional backing for AI infrastructure at a time when data center electricity demand is projected to grow 300% over the next decade and account for around 38% of U.S. net electricity consumption through 2037, according to the National Electrical Manufacturers Association.

   The AI Infrastructure Partnership has a remit to support up to $100 billion（約16兆円） in total investments and has added xAI, Singapore's Temasek, and the Kuwait Investment Authority as members since its launch. Aligned will remain Dallas-headquartered under CEO Andrew Schaap, with the new owners providing strategic AI insights and long-term capital while the company maintains operational independence.

2. 2

   Salesforce adds carbon, energy metrics to AI model cards

   Salesforce has begun publishing energy consumption and carbon emissions data alongside its AI model cards — fact sheets that describe how models were trained and perform. As of June, the company started including these metrics for models it trained directly, such as first name match and account match models. Model cards, sometimes called AI "nutritional labels," typically cover training methods, performance, applications, and demographic factors, but Salesforce is unique in adding environmental impact data. Organizations scaling AI are asking how to deploy it responsibly, including understanding its environmental impact alongside performance, cost, and business value. Salesforce has received positive feedback on the transparency, with customers hoping other model providers will follow. The company's approach addresses a gap: while training impact is straightforward to measure, inference impact — when AI is applied to a specific task — varies by location and is harder to estimate, yet Salesforce's calculations include that phase.

   Salesforce plans to add environmental metrics to additional models as measurement techniques become more straightforward. The company's AI Energy Score, calculated using a resource published by open source firm Hugging Face, analyzes hardware, data center location, run times, and other factors. The sustainability team collaborated with Salesforce's office of ethical and humane AI use to create the data.

3. 3

   Vertiv expands AI cooling with Strategic Thermal Labs buy, Italy capacity

   Vertiv Holdings Co acquired Strategic Thermal Labs and committed to doubling chiller production capacity at its Tognana campus in Italy, while also building a large-scale AI workload testing lab. These moves deepen the company's engineering capability in high-density liquid cooling for power-hungry AI data centers. Vertiv's investment thesis rests on AI-driven, high-density data centers requiring specialized cooling and power infrastructure, and the company's ability to scale while protecting margins. The expansion directly supports that belief, though it also raises execution risk around integrating new technology, ramping new facilities, and potential cost overruns. The company's narrative projects $22.2 billion（約3.6兆円） revenue and $4.2 billion（約6700億円） earnings by 2029, requiring 27.0% yearly revenue growth and a $2.6 billion（約4200億円） earnings increase from $1.6 billion（約2600億円） today.

   The moves tie into near-term catalysts of growing AI data center demand and complex liquid cooling requirements, but investors should monitor supply chain reconfiguration, EMEA execution, and the risk that large cloud customers could eventually bring more cooling and power solutions in-house. Some analysts were already cautious, assuming Vertiv might need to earn about US$4.1 billion（約6600億円） by 2029 to justify a lower price target.

4. 4

   Shield AI partners with Honeywell on autonomous aircraft software

   Shield AI, a San Diego startup, signed a memorandum of understanding with Honeywell to combine Honeywell's certified avionics, navigation and sensing technologies with Shield AI's Hivemind autonomy software. The partnership comes days after Shield AI acquired Aechelon, a California software company that trains military pilots. Proving that AI-powered aircraft can be trusted by regulators, military operators and commercial customers is one of autonomous aviation's biggest hurdles. By pairing certified flight systems with autonomy software, the two companies aim to engineer trust into the architecture—a shift that reflects how aerospace software companies are increasingly partnering with established hardware providers to credibly deploy AI in aviation.

   The companies will jointly pursue unmanned aircraft opportunities for defense customers worldwide and explore future commercial transportation applications. Shield AI's private share valuation is estimated at $13.2 billion（約2.1兆円） by Forge Global, slightly above its March fundraising valuation.

5. 5

   Pfizer scientist wins Select Science award for oncology research video

   A Pfizer scientist received the Select Science Award for Video of the Year for work on oncology research and innovation. The scientist highlighted the importance of quality data and analysis in cancer research, even as AI and machine learning tools become more prevalent in the field. Cancer rates are rising and affecting younger people, making oncology research a critical endeavor. The scientist emphasized that while AI and machine learning can accelerate research, they cannot replace the need for high-quality data acquisition and rigorous analysis—a reality that underpins the value of collaboration between researchers and analytical equipment makers like Shimadzu.

   The collaboration between Pfizer and Shimadzu, which has spanned 12–13 years and intensified over the last five to six years around newer chromatography and SFC (supercritical fluid chromatography) techniques, demonstrates how equipment innovation supports drug development pipelines in oncology.

6. 6

   Schwab Q2 earnings beat, raises 2026 outlook to 17.5–18.5% revenue growth

   Charles Schwab reported second-quarter adjusted earnings per share of $1.62, beating the Zacks Consensus Estimate of $1.53, with revenues of $7.07 billion（約1.1兆円） exceeding the estimate of $6.89 billion（約1.1兆円）. The company added 1.4 million brokerage accounts and generated $120 billion（約19兆円） in core net new assets during the quarter. Management raised its full-year 2026 revenue growth outlook to 17.5–18.5% and expects net interest margin expansion to 3–3.10%. The stronger-than-expected results reflect Schwab's ability to deepen client relationships across investing, wealth, banking and lending on a single platform. Managed investing net flows increased 53% year over year, and bank lending balances reached $67 billion（約11兆円）, up 33%, signaling diversified revenue growth beyond trading. CEO Richard Wurster emphasized that only a small portion of retail clients currently use fee-based advice solutions, positioning wealth services as a significant future growth opportunity.

   Schwab is advancing AI-driven capabilities across its platform, including Portfolio Insights and Schwab Assistant, with developer productivity improving 15–20% over the past year. The company also highlighted Schwab Crypto, with transfer capabilities expected to enter pilot testing, and the Forge acquisition, intended to expand private market access. Management raised the full-year daily average trade assumption to 10.6 million trades.