Morgan Stanley

Anthropic announced a $100 million investment, or roughly 3.6 billion baht, to establish an artificial intelligence training institution called the Claude Frontier Academy, with the goal of cultivating 10,000 frontline hands-on engineers from companies in its Claude Partner Network and having them complete training by the end of 2027. The first group of participants includes software engineers from Accenture, Morgan Stanley, and Novo Nordisk. The curriculum begins with intensive simulations of enterprise technology deployments before moving into a residency program modeled on the teaching approach used in medical schools, and the first cohort of engineers is expected to receive official certification in early 2027. The move comes amid expectations that Anthropic will go public this year, with a target valuation of as much as $2 trillion, after Reuters previously reported, citing a leaked draft prospectus, that the company had revenue of nearly $4.6 billion last year but an operating loss of more than $8 billion.

Morgan Stanley cut its Nike price target to 27 from 31, with analyst Alex Straton saying the company's first quarter is likely this year's high watermark rather than the low. Straton, who went underweight on the stock late last year, said Nike's guidance embeds a high single digit sales decline and that reaching full-year targets requires both top-line and earnings trends to degrade from here. She attributed the pressure to excess inventory in North America and a similar dynamic in China, the two biggest sportswear markets in the world, compounded by a challenging macro backdrop. Straton flagged three problem areas for the company: Sportswear, which makes up almost two-thirds of the portfolio and roughly 30 billion dollars in revenue, the Jordan brand, and a material change in the China outlook. She said she has not seen the bottom in negative revisions and that Nike may face a shrink-to-grow story, with further de-rating possible if the company settles into a lower-growth, lower-margin multiple. Nike holds an investor day in November, where Straton said the company needs to clarify the right revenue size for Sportswear, Jordan and China before investors can gauge when growth and historical margins might return.

A new study from the AI company Anthropic suggests it will take 40 years for robots to replace just 10% of job tasks if robot prices decline along past trends, a timeline that clashes with the more bullish scenarios of Nvidia and Tesla. At CES in January 2026, Nvidia CEO Jensen Huang predicted a coming "ChatGPT moment" for robotics in the form of agents, self-driving cars, and humanoid robots, while Tesla CEO Elon Musk has said the company's Optimus robot will go on sale as early as next year and envisions a future where every household has a robot. A Morgan Stanley research report from 2025 forecasts that the market for humanoid robots could surpass $5 trillion by 2050, with adoption likely to accelerate in the late 2030s on improved tech, a friendlier regulatory environment, and popular support. The Anthropic report argues that beyond price, factors including capabilities, preferences, and regulations pose further barriers to robot automation, and Apollo chief economist Torsten Sløk echoed that "even under aggressive projections, widespread physical labor displacement will take decades." The disparity underscores the tension between the rapid pace of software development and the slow progress of hardware, leaving open whether robots will become commoditized gadgets or premium products with constrained supply.

Microsoft shares closed the July-September quarter up 37.5%, the biggest quarterly gain since 1998, adding $1 trillion to the company's market capitalization, according to Bloomberg. The main driver was the late-July earnings report, in which the Cloud business grew at its fastest pace in four years on strong demand for AI services, sending the stock up 16% in a single day, its best daily gain since October 2008, and adding about $450 billion to the company's market value. Among the big technology companies pouring money into AI, including Alphabet, Amazon and Meta Platforms, Microsoft is the only one whose free cash flow has not turned negative on an annualized basis. Meanwhile, of the 72 analysts tracked by Bloomberg, only 3 do not have a buy rating, and none recommend selling, with the average price target implying the stock could rise another 11% over the next 12 months. Adam Wood of Morgan Stanley estimates the stock could deliver a total return of about 20% over the coming year, and Wall Street expects revenue to rise about 18% in fiscal 2027 before accelerating to about 21% in fiscal 2029. Earnings per share are expected to rise about 11% in the current fiscal year, slowing from nearly 32% in fiscal 2026, before reaccelerating to nearly 19% in fiscal 2028 and about 21% in fiscal 2029.

Morgan Stanley is reportedly exploring a potential merger or combination with UBS AG under pressure from Swiss regulators, who are said to be urging UBS to reinforce its capital position and prompting discussions with several large foreign banks. Any Morgan Stanley and UBS tie-up would rank among the largest cross-border banking deals and could reshape global investment banking. Morgan Stanley, a US-based capital markets group with a market cap of about $304.1b, runs a securities and advisory franchise spanning the Americas, Europe, Asia, the Middle East and Africa, and a UBS combination would add a deep European and cross-border wealth footprint that complements its existing US$10t in client assets. The report cuts both ways for the bank's Wealth Scale narrative: it reinforces the case for using surplus capital and regulatory headroom on selective acquisitions rather than only buybacks and organic projects, but a bigger, more complex balance sheet could pressure capital strength, invite fresh regulatory demands and distract from adviser retention amid poaching risk and fee pressure from passive products.

Morgan Stanley Investment Management announced the upcoming liquidation of the Calvert US Large-Cap Diversity, Equity and Inclusion Index ETF, ticker CDEI, a series of Morgan Stanley ETF Trust, after the Trust's Board approved a Plan of Liquidation. Under the plan, the fund's assets will be liquidated, known or reasonably ascertainable liabilities satisfied or provided for, remaining proceeds distributed to shareholders, and all issued and outstanding shares redeemed, with the liquidation expected on or about November 2, 2026. Effective at the close of business on October 23, 2026, the fund will stop accepting orders for the purchase of Creation Units, and that date is expected to be its last full day of trading on NYSE Arca, where trading is expected to halt at market close. Shareholders may sell their shares on NYSE Arca until market close on October 23, 2026, or remain invested, but between that close and the liquidation date the shares will not trade on NYSE Arca and there is no assurance a market will exist for their purchase or sale. Before the liquidation date the fund will transition its assets to cash and cash equivalents, and during that period it may no longer pursue its investment objective or be managed consistent with its stated strategies, which is likely to impact performance. Shareholders still holding shares on the liquidation date will receive a liquidating distribution that may be greater or less than what they might receive by selling on NYSE Arca before market close on October 23, 2026, and may bear increased brokerage and other transaction expenses; sales or liquidations will generally be taxable events outside tax-deferred accounts.

Morgan Stanley sees roughly 50% upside in Oracle shares but is not recommending investors buy them, with analyst Sanjit Singh maintaining an Equal Weight rating and a $210 price target. The central concern is what Oracle gets back from its massive AI infrastructure spending, even as Oracle Cloud Infrastructure revenue rose 121% year over year in the fiscal first quarter. Singh said he wants to see further gross margin stabilization, though he noted promising signs including greater OCI capacity, higher renewal pricing and increasing profits, and he liked the company's reaffirmed annual capital expenditure projection of $90 billion to $95 billion. Oracle's financial analyst day on Oct. 28 is the next key driver, which Singh called a good starting point to see whether Oracle can clarify infrastructure deployment, profitability and application patterns. By contrast, Morgan Stanley analyst Joseph Moore keeps an Overweight rating on Broadcom with a $505 price target, believing it will eventually capture more than 80% of serviceable markets with its custom-chip projects, even though he thinks 2027 AI revenue forecasts are too high.

The Chinese government announced a new round of stimulus measures aimed at boosting lending liquidity for targeted sectors and subsidising housing loans, after markets closed on 29 September. It is the first major stimulus package since September 2024. The Ministry of Finance will allocate housing loan subsidies to eligible homebuyers starting 1 October. First-time homebuyers will receive a subsidy equivalent to 1% per year of their mortgage principal for a maximum of five years. Meanwhile, the central bank expanded the quota of its relending facility for lending to targeted economic sectors such as infrastructure, technology and small businesses by a further 200 billion yuan, or 28 billion dollars, bringing the total to 1.4 trillion yuan. It also raised the share of qualifying loans that can draw on the facility to 100% from 60%, and cut the one-year interest rate on the facility by 0.25% to 1.5%. China has set this year's economic growth target at 4.5-5%, while second-quarter growth slowed to 4.3%, below target. Duncan Rigley, chief China economist at Pantheon Macroeconomics, said the measures would likely lift Chinese growth to the lower end of the target range but would not fix structural imbalances, including sluggish domestic demand and heavy reliance on exports. Analysts at Morgan Stanley said the modest stimulus would not be enough to change the scale of household and local government debt in China, and that the latest property support measures would benefit lower-tier cities but would not comprehensively fix or rescue the real estate market.

Morgan Stanley analyst Andrew Percoco estimates autonomous Tesla Semis could generate roughly $17 billion in annual software revenue by 2040, a scenario that assumes about 82,000 Semis operating by then and a 13.5% share of the autonomous trucking market. At that scale, the firm estimates Semi software could produce about $7.5 billion in EBIT, roughly 10% upside to its base case. Percoco estimates Tesla could charge roughly $0.85 to $1 per mile for autonomous Semi software, which at about 18,000 miles a month works out to approximately $12,000 to $18,000 in monthly software revenue per vehicle, far above the roughly $100 per month Tesla charges consumers for Full Self-Driving. Morgan Stanley views its 82,000-vehicle assumption as relatively modest against Tesla's manufacturing ambitions, with the company's 1.8 million-square-foot Semi factory in Sparks, Nevada, designed to eventually produce about 50,000 trucks annually. Early Semi customers include PepsiCo, DHL and US Foods, while a recently announced shipper coalition has ordered 2,500 trucks; the $17 billion estimate covers software rather than Semi vehicle sales and excludes potential additional revenue from charging.

Tesla opened its dedicated Sparks, Nevada factory and began customer deliveries of its long-delayed Semi truck, a plant designed to eventually produce 50,000 Semis annually. The truck is offered with claimed ranges of 325 miles and 500 miles for the long-range version, and early customers include PepsiCo, DHL and US Foods, while a newly announced 2,500-truck order from a shipper coalition adds to potential demand. Morgan Stanley analyst Andrew Percoco wrote in a new note that Tesla can generate roughly $12,000 to $18,000 per month from its autonomous trucking software, based on a $0.85 to $1 per mile subscription fee across 18,000 miles driven per month, compared with about $100 per month for full self-driving on consumer vehicles today. Percoco calculated that if Tesla reaches 82,000 Semis on the road by 2040, a 13.5% share of the autonomous addressable market, that would represent $17 billion of software revenue and about $7.5 billion of incremental EBIT, a 10% upside to Morgan Stanley's base case, and he noted this excludes the upfront vehicle sale and incremental charging revenue. He added that Tesla originally outlined a manufacturing target of 50,000 Semi trucks per year, so assuming 80,000 will be deployed in total over the next roughly 15 years is a conservative starting point.

Morgan Stanley strategist Michael Wilson said the gap between the S&P 500's performance and the market's weak breadth needs to close, and he expects the two to meet somewhere in the middle over the next month unless bond market volatility eases. In a note on Monday, Wilson wrote that the firm leans toward a meeting in the middle if bond volatility does not calm down. Morgan Stanley still sees the market in a classic mid-cycle phase, with strong earnings offsetting lower valuations and higher-quality stocks taking the lead. Wilson noted that more than half of Russell 3000 stocks have fallen at least 20% since June, while the S&P 500's P/E ratio is back near 19 times, its March low, and earnings growth for the median stock remains in the mid-teens. He added that market breadth improved through most of the summer even as energy prices and yields rose, and that the recent narrowing in breadth occurred after Jackson Hole as the market priced a more hawkish Fed path. If bond volatility persists, Morgan Stanley expects breadth and index prices to meet in the middle over the next month before a strong finish to the year; if volatility eases sooner, the bank believes breadth should catch up with the index, lifting both. Wilson continues to favor large-cap quality stocks, especially asset-light companies with rising earnings estimates, and would add riskier stocks over the next month if the index falls. Morgan Stanley also said companies making heavy use of AI are seeing stronger margins and earnings, and that consensus assumes those benefits fade in later years, leaving room for further upside.

Dell Technologies shares climbed 4% Friday after Morgan Stanley raised the probability attached to its $756 bull-case scenario following a meeting with Chief Operating Officer Jeff Clarke, while leaving its Neutral rating and $511 target unchanged. The shift centers on whether AI spending represents another hardware replacement cycle or a broader expansion in infrastructure demand. Clarke told the brokerage that AI agents could drive inference-token usage 87 times higher by 2030, potentially creating additional demand for computing capacity. Morgan Stanley's unchanged base-case valuation means the note did not alter its formal recommendation, but the higher probability assigned to the more optimistic scenario gave investors another data point as they assess whether Dell's AI infrastructure growth can persist beyond the current spending cycle.

Morgan Stanley has formally changed its Bank of England call to two rate hikes, in November and February, pointing to fiscal policy as the key potential catalyst for further tightening beyond that. Economists led by Bruna Skarica said they see the September meeting and Deputy Governor Lombardelli's remarks as "quite balanced," with the Monetary Policy Committee keen to stress that both the November hike and the broader path ahead remain uncertain. The bank said newsflow in commodity markets "is not improving," making it harder to justify a prolonged pause as its central scenario given recent developments in the Middle East, and it noted its call still looks dovish relative to roughly 100 basis points currently priced into the front end of the U.K. yield curve. Morgan Stanley estimates current fiscal headroom at approximately £8 billion, and said a £5 billion uplift to day-to-day spending to account for higher inflation could further reduce that buffer, meaning between £10 billion and £15 billion of revenue-raising measures may be needed to lift the budget surplus toward £15 billion without additional spending cuts. The bank also updated its gilt yield forecasts to reflect the new Bank Rate path, projecting two 25-basis-point hikes in November and February, and said attention is expected to increasingly shift toward the Budget, particularly the scale and front-loading of fiscal consolidation and its implications for near-term gilt supply.

Morgan Stanley on the 25th revised its UK interest rate outlook, predicting that the Bank of England will raise rates by 25 basis points each in November and February next year. Previously it had expected no rate hikes for the time being. In a research note, analysts said they had formally revised their Bank of England forecast to two rate hikes per quarter, and expressed the view that fiscal policy would be the main potential catalyst for further tightening. They also noted that growth is expected to slow around the year-end and New Year period, but that near-term risks are balanced given the resilience of global growth.

Bualuang Securities says DELTA's growth outlook from AI and data centers has strengthened after Morgan Stanley raised its power assumptions for the Vera Rubin rack to 234kW from 149kW, and for Rubin Ultra to 600kW from 415kW, reflecting power demand that comes not only from GPUs but also from memory, networking, power delivery systems and cooling across the entire rack. As a result, Morgan Stanley now estimates global server and rack power demand in 2026 to 2028 at 18GW, 35GW and 52GW, compared with Bualuang's assumptions of 15GW, 20GW and 25GW respectively. Even after deducting US electricity constraints, this still equates to global installations of about 28GW per year, above Bualuang's assumption of 20GW per year. For DELTA, the higher power usage per rack supports both the value of power delivery systems and cooling systems per rack, while Morgan Stanley has raised its liquid cooling capex assumption by 30% and sees cooling as an increasingly essential infrastructure. In addition, the 800VDC architecture improves system efficiency by about 5%, cuts copper usage by about 45% and supports racks drawing more than 1MW. In the short term, component constraints look set to ease gradually in the third quarter of 2026, with Delta Taiwan's power electronics sales in July up 59% year on year and 17% month on month, before falling 9% month on month in August but still 40% above the same period last year. DELTA's third-quarter 2026 sales are expected at 2.2 billion US dollars, up 34% year on year and 10% quarter on quarter, with a gross margin of 29.2% and core profit of about 9 billion baht, up 21% year on year and 60% quarter on quarter, before the fourth quarter of 2026 has a chance to set a new record high on deliveries of orders backlogged from the second and third quarters of 2026. Bualuang maintains its Buy recommendation with a target price of 330 baht and keeps its weighting on the electronics sector at market weight.

Morgan Stanley has put a number on one of the AI boom's most awkward constraints: the United States could face a roughly 33-gigawatt power shortfall through 2028 even after accounting for onsite generation and other accelerated power solutions. The bank's September analysis points toward behind-the-meter generation and equipment suppliers as part of the bridge, with Bloom Energy Corporation able to place fuel-cell generation close to the load and GE Vernova Inc. selling the turbines and grid equipment needed to add large blocks of reliable power. Bloom's Q2 revenue topped $1 billion and management has guided to roughly $3.9 billion to $4.2 billion for the year, while GE Vernova's gas-turbine and electrification backlogs give it exposure to multi-year power investment rather than a single data-center technology. Hedge-fund ownership in GE Vernova fell to 106 funds in Q2 from 118 in Q1, and Insider Monkey counted 116 hedge funds holding Bloom in Q2 2026, up from 91 in Q1. A 33-gigawatt deficit is large enough for both approaches to work, but the better operating outcome will depend on whether AI campuses prioritize speed, long-run power cost, or both.

Brookfield Asset Management's retail arm GGP is locking in an $800 million commercial mortgage-backed security refinancing for Oakbrook Center, one of the last large, functioning malls in suburban Chicago. A syndicate comprising Morgan Stanley, Bank of America, Citibank, Goldman Sachs and Wells Fargo are reportedly the originators and sellers of the $800 million loan, according to Bisnow. The CMBS loan is set to repay a previous $700 million CMBS loan tied to the Oakbrook Center at 100 Oakbrook Center in the suburb of Oak Brook, as well as pay off $30 million in early payoff penalties, with the remainder of the cash going to $5 million in closing costs and returning $65 million of equity to the sponsors. The loan carries a five-year interest-only term and an assumed rate of 5.9 percent, and in July 2026 the mall was 94 percent leased with around 160 unique tenants. Oakbrook Center is the second-largest shopping mall in the Chicago metro and not only came out of the pandemic alive, but thriving.

Morgan Stanley has warned that the S&P 500 could fall 7% to as low as 7,100 before the bull market resumes into year end, citing tightening financial conditions and materially higher energy prices. The firm said bond yields near 5% would make equities less attractive to investors, while rising oil prices would seep through much of the economy. Diesel prices are already feeding persistent inflation, and with 70% of U.S. freight moving by truck, that pressure is amplified. AI stocks face twin risks: a single major security breach or a pullback in corporate AI spending could erase the sector's market momentum. The firm said inflation is a more likely trigger for a selloff than an AI flameout, since capital for data centers remains available and AI demand momentum is unlikely to disappear in a quarter.

Technology companies have provided as much as $300 billion in guarantees for AI data centers and chips over the past year, according to the Financial Times, creating a financing structure that could change how investors assess the risks behind the artificial intelligence spending boom. The guarantees establish minimum future values for computing equipment or data centers owned by special-purpose entities, and if those assets later sell or lease for less than the guaranteed amount, the technology company could be responsible for part of the shortfall. Broadcom has provided about $29 billion in backing for a vehicle that will purchase chips and lease them to Anthropic, while Nvidia has agreed to support as much as 25% of certain infrastructure financing packages and provided $105 billion in guarantees tied to an Ohio data center campus for OpenAI. Meta has also provided a $28 billion guarantee for a data center venture with Blue Owl, and Morgan Stanley estimates seven major cloud and chip companies have accumulated more than $3.1 trillion in off-balance-sheet commitments and other credit support. Analysts generally view the underlying assets as sufficient collateral so far, but the equation could change if AI demand slows, equipment becomes obsolete faster or data-center capacity exceeds customer needs, leaving investors to watch whether AI infrastructure spending generates enough cash flow to support the financing structures behind the boom.