Meta’s ad engine is converting AI investment into measurable performance gains, while Muse offers a potentially large new monetization surface.
The AI buildout risks overwhelming cash generation and margins; dilution, uncertain returns, and limited 2027 commitments could cap the stock’s multiple.
The stock held up well, indicating resilience.
META +2.4% held up well while GOOGL lagged. · OpenAI told Alex Kantrowitz its internal data shows ChatGPT search volume is already about a fifth of Google’s:
+106% y/y, while aggregate cloud growth accelerated to 57% from 48% last quarter. The debate has shifted from demand to ROI/financing as CY27 capex estimates rise toward ~$1.5T including SPCX/CRWV, with combined hyperscaler + META + ORCL FCF not expected to...
S&P affirmed Meta’s AA− rating but cut its outlook to negative, forecasting significant capital expenditures and free operating cash-flow deficits through 2028.
Data-center leases could push leverage above 1x in 2027, and the forecast excludes Meta’s planned $60B at-the-market equity program.
The company demonstrates superior ROI and ad performance compared to peers, proving that LLMs can materially improve advertising efficiency.
META +1% getting some love after the GOOGL miss. Neoclouds lower with NBIS -7% as some speculating the MSFT 3p capacity delay is them.
The table shows that while GOOGL has a significantly higher headcount, META consistently demonstrates higher LTM Adj.
AVGO: KeyBanc confirms Anthropic ASIC win, raises OpenAI volumes and says META Apollo remains in play
The company's $25 billion stock sale was significantly oversubscribed, indicating strong investor demand.
DA Davidson upgraded to Buy, arguing AVGO is pricing in near-zero odds of an Anthropic IPO/raise or more OpenAI funding. KeyBanc said AVGO has officially won Anthropic’s ASIC (late-2028/early-2029, ~5-10M lifetime units), while META’s Apollo training ASIC r...
AMD also said they will put $1B into META's equity offering and confirmed its first Helios racks are installed at META and in testing, while META shared a release on training production recommendation models across tens of thousands of MI350X GPUs:
KeyBanc says AVGO has officially won Anthropic’s ASIC, a potentially sizable late- 2028/early-2029 program with ~5–10M lifetime units, while correcting its prior view that META’s Apollo training ASIC had gone to MTK; that award remains open and AVGO is stil...
The company's stock slipped despite its stock sale being oversubscribed.
The company's equity plan, including a significant raise, provides a straightforward path to capital, which is positive for the AI semis trade.
Arun Dhar asked in Slack here “why isn’t it reacting like a clearing event?” …DR pushed back that MSFT and AMZN capex is accelerating >20pp each while META is unlikely to accelerate even ~10pp, and he expects more issuance for 2028. ALC‘s math here: ~$...
The company's next individual-plaintiff bellwether trial for mental-health cases could result in significant damages, with a prior bellwether producing a $4.2M award that, if extrapolated, could imply substantial future settlement economics.
A separate $25B New Mexico judgment against the company is likely to be materially reduced on appeal.
The company's equity raise, while substantial, is unlikely to accelerate capex as much as other hyperscalers, suggesting a less aggressive investment posture.
META -1.9% launched the first $35B of a ~$95B equity plan: $22.5B of Class A via forward sales plus ~$12.5B of mandatory convertible preferred now, then a ~$60B Class C ATM once the Class C trades in 2027, with NVDA, AMD and AVGO indicating interest in up t...
The company's third-quarter revenue and EPS, excluding a settlement, surpassed Street expectations, with fourth-quarter guidance also slightly exceeding targets, and management highlighted the commercialization potential of Muse.
Bulls were encouraged by the two-year acceleration in U.S. and Canada (UCAN) revenue growth, which showed high incremental gains, and the fourth-quarter trends implied continued acceleration.
The stock's risk/reward near $650 is attractive, making it a long-term holding despite near-term digestion.
The narrowing of the capex range by raising only the low end was a slight positive for long-term investors, as it came in slightly below buyside expectations for AI semi bulls.
Advertising revenue growth accelerated in the U.S. and Canada, with global pricing increases more than offsetting impressions growth, supported by improved ad performance and macro conditions.
Evidence continues to build that AI is improving the existing advertising business, with advertisers adopting Business Assistant recommendations generating more conversions and increasing spending, and generative creative tools seeing widespread use.
Muse adoption is strong, with over 10 million people in the U.S. and Canada setting up agents, and a significant portion using Muse daily, with shopping being a leading use case involving transactions averaging over $200.
Management intends to monetize Muse earlier than its historical playbook due to substantial serving costs and revenue opportunities, with planned streams including transaction fees, subscriptions, and connections to content and advertising experiences.
The company has substantial distribution for business agents, with at least 500 million self-identified small businesses using its apps and 3.64 billion daily consumers, creating a significant opportunity for connecting consumer and business agents.
Other revenue for Family of Apps (FoA) increased significantly, driven by WhatsApp paid messaging and subscriptions, with partnership ads exceeding a substantial annual revenue run rate and Meta One surpassing 15 million subscriptions and trials.
The company is targeting a significant improvement in Muse compute cost-efficiency over the next year through smaller models, distillation, inference optimization, and better utilization of spare capacity.
Operating income, excluding the settlement, was significantly above Street expectations, despite a decline in margin due to increased underlying expenses.
The underlying expense range for FY26, excluding the AG charge, remained effectively unchanged, providing a slight relief.
Third-quarter capex was below expectations, which helped free cash flow come in better than the Street's negative estimate.
Concerns about 2027 EPS growth are weighing on the stock because management only committed to "Strong revenue growth" without an EBIT commitment, leading to uncertainty about potential numbers.
Equity raise and capex concerns are impacting the stock, as a 14C filing indicated a need to raise "hundreds of billions of dollars over several years" starting as early as fall 2026, with management recognizing the importance of securing capital before competitors saturate funding options.
Bears argue that catalysts are in the rearview mirror, Muse engagement has been mixed, carries different variable costs than ads, will take time to monetize, and near-term catalysts are no longer present.
Buyside numbers for 2027 EPS are not increasing and remain around $34-35, with bears arguing that EPS growth will be minimal in 2027, spending will remain elevated, and an equity raise is an overhang, justifying a sub-20x multiple and a target of $650.
The stock likely needs to digest recent developments and may shift into a funding short camp near-term unless there is a change in narrative, more concrete signs of Muse monetization, or the equity raise overhang is lifted.
The implied fourth-quarter capex, based on the annual guide, is mathematically high, suggesting a significant increase from the third quarter.
While the company targets significant compute cost-efficiency improvements, management explicitly expects total serving costs to rise as adoption expands, indicating that efficiency gains will not necessarily lead to a decline in aggregate spending.
Operating margin declined year-over-year as underlying expenses increased faster than revenue growth, with infrastructure, depreciation, cloud, and third-party token costs absorbing advertising gains, and labor efficiency not offsetting the broader investment ramp.
For 2027, management expressed confidence in strong overall revenue growth but provided no numerical guidance for revenue, expenses, or capex, and made no commitment to operating income growth, creating uncertainty.
META -3% down after being up 6% last night after some uninspiring quotes from their 14C filing keep equity raise concerns front and center while ‘27 margin concerns have buyside less confident in ‘27 upside, LRCX/NOK trading down on mixed prints, IBM -8% fa...
The company's equity raise fears put financing for AI infrastructure back in focus, causing the stock to round trip its 6% pop.
The company's stock declined due to funding and 2027 margin concerns, as management only committed to strong revenue growth for 2027 without an EBIT commitment.
The 10-Q revealed substantial uncommenced leases and non-cancelable commitments, totaling approximately $810 billion, with a significant portion due in 2027.
The 14C filing raised concerns regarding funding.
Bears project a low-30s EPS at a sub-20x multiple, leading to a stock price below $650.
Some of it due to META’s -7% equity raise fears putting financing the infra trade back in focus and causing the stock to round trip its 6% pop; semicap moved lower on LRCX’s -12% guide; NOK -15% closed at lows after its mixed print despite a sizeable EBIT b...
The company had a solid print post-close, with its stock rising in after-hours trading.
Revenue exceeded both Street and buyside expectations, and the top end of the fourth-quarter guidance surpassed the ~$75 billion target.
Third-quarter revenue came in above the top end of the guidance, primarily driven by strong pricing, with price per ad increasing by 14% and impressions by 10%.
The fiscal year 2026 expense guide, excluding a $9.9 billion charge, remained unchanged at $165-$169 billion, which was a slight relief to investors.
Capital expenditure guidance was narrowed by raising only the low end, resulting in a midpoint of $140 billion, which was a small positive for META longs as it was slightly below the buyside's expectation of $140-$145 billion.
Reuters reported a New Mexico jury ordered the company to pay $25 billion over privacy claims.
Semafor reported that the company plans nonvoting shares and more stock sales to fund AI initiatives.
Travel sold off a day after outperforming on Jefferies’s better checks: EXPE -5%; BKNG -3.4%; ABNB -1.8%. My guess is maybe some selling ahead of the META print as investors remain concerned on the the agent overhang: a check we heard found LLMs send NYC ho...
Investors remain prepared for a potential equity raise, with the company creating non-voting Class C stock effective January 19, 2027, through a 1-for-1 stock dividend, with capital raising listed as a primary reason.
The narrowed capital expenditure guidance, while a positive for META longs, was a slight disappointment for AI semi bulls as it was a little short of their $140-$145 billion expectation.
The unchanged ex-charge expense guide implies fourth-quarter expenses of approximately $48-$52 billion, an increase from $41.7 billion in the third quarter, which already represents an annualized run rate of approximately $200 billion before any 2027 growth.
Free cash flow was negative $2.1 billion on $34.9 billion of capital expenditure.
Edge indicated that while the macro environment softened slightly near the end of the quarter, the company finished in the upper end of its range.
Bernstein remains Outperform and expects Meta to exceed 3Q revenue guidance, supported by strong ad checks and AI-driven targeting/creative.
Muse adds meaningful optionality, with Raymond James' mid-case scenario projecting over $50B in annualized revenue and upside scenarios reaching over $150B in revenue.
Bernstein views Muse as strategically important enough to reorient Meta around a broader AI OS opportunity.
META is seeking another ~6.5GW by YE28, ORCL has resumed leasing with NVDA credit support which Cowen views as an indication of future RPO growth, and GOOGL/MSFT remain active. The bigger issue is execution: 2027 equipment slots are already oversold, build...
Raymond James sees a neutral setup, as strong advertising and Muse optionality are offset by higher 2027 expense and capital expenditure expectations from the buyside.
The buyside's expectations for 2027 expenses are closer to ~$210–220B (vs. Street $203B) and CapEx at ~$225–250B (vs. Street $198B), making the print's outcome dependent on whether spending commentary clears this higher bar.
Bernstein identifies 2027 spending as the key risk, noting that while Muse's early traction allows for more investment, OpEx/CapEx could exceed current expectations as products scale, cloud deals progress, and compute needs increase.
Bernstein cautions that early adoption and TAM enthusiasm for Muse still need to demonstrate durable monetization.
Bernstein anticipates a tougher 4Q comparison and potential FX pressure.
The company's stock declined ahead of its earnings print, with key debates centering on 2027 spend and a possible equity raise, as buyside estimates for 2027 expenses and capex are higher than Street consensus.
Wells Fargo has cut its FY27 EPS estimate to $31-32 on higher opex, projecting a trough earnings year.
Macro conditions were softer at the margin near the end of the quarter, although the company still finished in the upper end of its range.
If opex and capex come in high, the sell side's ~$34 for '27 is likely too high.
SE -9% the main laggard again on its cloud foray and META -2.3% into tomorrow’s print.
OpenAI, and META seeking another ~6.5GW by YE28. The bottleneck is shifting from demand to delivery: 2027 equipment slots are already oversold, Texas/Ohio permitting could delay capacity, and the US capacity contracted since 1Q25 implies ~$2.25T of building...
Stronger advertising performance and AI subscription growth are expected to drive earnings upside in the coming years.
DDOG +4% on a Bernstein upgrade; META +40bps lagging on mixed MS/3p notes ahead of earnings on Wednesday; AMZN +2.4% as MS makes it a top pick.
SemiAnalysis argues 2027 data-center growth is materially underestimated, forecasting ~50GW of new capacity in a single year with >24GW already in advanced construction across just 40 campuses and ~60GW under construction more broadly. The key leading indic...
META: MS says every 100M Muse DAUs could require 1–4GW of GPUs, keeping capex higher for longer
Sensor Tower data shows Muse had over 10M downloads in its first 35 days across the US and Canada, with 23% US 14-day retention, which is higher than Claude.
High GPU capacity requirements for AI agents suggest sustained high capital expenditures, and the Muse product currently faces engagement challenges.
Morgan Stanley flagged Muse's engagement problem, noting approximately 2.5M DAUs and only 20% 14-day retention.
Muse could require significant GPU capacity, with 100M DAUs potentially consuming 1-4GW and 500M DAUs using 50% of free token allowance requiring approximately 13GW, which is roughly all of Meta's modeled incremental 2027/28 capacity.
Positioning into the print has been trimmed due to equity-raise risk, indicating investor caution.
We don’t disagree and think stock was helped by META underperformance today as investors shuffled their preferences into the hyperscaler prints. The CFO's payback math helps bulls who argue servers pay back in <3 years on 5-6 year lives, so negative FCF is...
META -55bps lagged into Wednesday’s print with the morning notes mixed: MS raised numbers on stronger ads and AI subscriptions but flagged Muse’s engagement problem (~2.5M DAUs, ~20% 14-day retention) and said 100M Muse DAUs could consume ~1–4GW of GPU capa...
Oppenheimer raised 2028 revenue, EBIT, and EPS estimates, anticipating improved ad monetization.
Oppenheimer flagged a potential FY27 AI cost reset, estimating substantial serving costs for Muse that would only be partially offset by savings from eliminating Anthropic API spend.
The analyst cut 2027 operating income and EPS, and sees potential for an equity raise ahead of Anthropic’s IPO.
Lots of debate about how much distillation/META/MSFT headwinds affected the Q3 number (along with slower summer seasonal months + Astra’s release in September) with a rough estimate probably around $4-7B, which would have put ARR in mid to high 70s range. N...
Instagram's global DAU accelerated to +6.5% year-over-year, its fastest growth since May '24, indicating strong user engagement.
TikTok is growing faster than Instagram in terms of time spent (+17%) and DAU (+10.5%), indicating competitive pressure.
The stock slipped after reports that Senate Democrats are investigating its use of R&D credits for data centers, raising potential regulatory scrutiny.
Evercore ISI believes Muse is scaling faster than ChatGPT, having achieved 5 million U.S. downloads in 22 days compared to ChatGPT's 56 days, and reached approximately 2.1 million daily active users by October 12, suggesting a path to 100 million users within 12 months.
Evercore ISI estimates a significant revenue opportunity of $50–100 billion over 3–5 years for Muse across subscriptions, ads, and transactions, with even a small percentage of Meta's daily users converting to paid Muse generating substantial revenue.
Pivotal raised its price target, noting that Muse outpaced ChatGPT's early mobile launch in the U.S./Canada iOS market and that new Wonder model benchmarks are competitive with frontier peers, improving confidence in Meta's AI returns.
Pivotal incorporated Muse and enterprise AI into its forecasts, projecting significant revenue contributions by 2030, and increased its terminal EBITDA multiple, reflecting strengthening confidence in Meta's AI monetization and return on invested capital.
Mark Zuckerberg's comments suggest that AI agents like Muse could dramatically increase the total addressable market for commerce by creating new users and workflows, rather than just improving conversion rates for existing ones.
Meta is an initial customer for AMD's Helios, with systems installed and running, indicating early adoption of AMD's new GPU architecture.
We’re settling around a $78–80B September ARR bogey versus $65B at July-end based on our polling and conversations, though there’s plenty of debate in both directions. Bears see $80B as a stretch given potential headwinds from META/MSFT curtailing usage, Op...
The CEO's comments about a "big model" coming soon, potentially related to Prometheus, suggest positive developments in AI.
The release of the Muse Wonder model strengthens the bull case that the company can generate new revenue streams through model APIs and direct enterprise sales.
While Zuck said they weren’t slowing spending, BTJ explained the bearish readthrough from Muse Wonder in Slack here: the frontier was “an effective duopoly that required infinite external capital raising”; if it’s now an oligopoly with META (and potent...
The company's commitment to high spending levels and the potential for significant capital expenditure increases raise concerns about future profitability.
CPM growth accelerated to +47% in September from +33% in August.
META CPM growth accelerated to +47% in September from +33% in August, while GOOGL’s share of AI Mode/Overview citations rose to 21.4%, +290bps m/m/+540bps since April. Ad checks remain strong into 4Q, with NFLX/ROKU/YouTube/RDDT standouts and TTD/SNAP weakest
The CEO pushed for the launch of Muse despite internal safety concerns, including an unauthorized password change, after seeing traction for a competing agent.
Muse has reached 6.6M downloads and 1.8M daily active users despite privacy lapses, with the company spending millions to market its Jolly avatar.
Daily active users on Muse continue to climb, peaking at 2.182 million, demonstrating strong user engagement and growth.
Despite challenges, 40% of advertisers expect Muse to increase their spending on the platform over the next 12 months, indicating confidence in its advertising potential.
The company remains a favored long position as Muse and Business Agents are expected to expand AI monetization opportunities.
Internet: Evercore reshuffles top longs to AMZN/META/SPOT; TAP Outperform RDDT, Underperform BKNG
Evercore moves AMZN to #1 Large Cap Long on AWS accelerating to +37% y/y, ~39% margins and ~$500B backlog, with PT $355/>40% upside; META falls to #2 but remains favored as Muse/Business Agents expand AI monetization, PT $860; and SPOT replaces DASH at #3 a...
The agent access bottleneck, with many sites blocking or restricting Muse, could delay the company's opportunity in agentic-commerce take-rates.
Florida is seeking immediate restrictions on the company's platforms, including removing accounts for children under 14, limiting teen usage, and disabling advertising for minors, which could impact user base and revenue.
Ad checks continue to improve, with Deutsche Bank noting an acceleration in growth of approximately 200 basis points.
The company is considered the 'king of the agents for now,' with its Muse product being very good at synthesis and superior for normie tasks, such as logging into utility websites and pulling property taxes.
Anthropic: The Information reported at 2:15pm that META and MSFT, two of its biggest corporate customers, are working to wean staff off Claude: MSFT cut its projected internal Claude spend to under $700M from at least $1B, steering employees to fewer Claude...
META +1.9% with ad checks still improving: DB had META growth accelerating ~200bps this morning (its PINS +4.2% checks improved too). Other checks in general have been positive here too. Main pushback from investors right now is on on costs, with 2027 expen...
The main concern from investors is rising costs, with 2027 expenses projected to be in the mid-single digits above Street expectations and capital expenditures exceeding $250 billion due to the Muse ramp.
Third-quarter agency checks indicate strong year-over-year spend growth, with impressions up and pricing at its strongest since Q1 2022, supporting the bull case for Q3 results at the high end of guidance.
Bears are concerned about the 2027 opex/capex reset, with Wells Fargo modeling significantly higher FY27 opex compared to Street estimates.
Ad checks continue to improve, with one firm noting accelerating growth of approximately 200 basis points.
The company is considered a leader in agents, with its Muse product being praised for its synthesis capabilities and integration with various accounts, making it a superior 'normie product' compared to competitors.
Anthropic: The Information reported at 2:15pm that META and MSFT, two of its biggest corporate customers, are working to wean staff off Claude: MSFT cut its projected internal Claude spend to under $700M from at least $1B, steering employees to fewer Claude...
META +1.9% with ad checks still improving: DB had META growth accelerating ~200bps this morning (its PINS +4.2% checks improved too). Other checks in general have been positive here too. Main pushback from investors right now is on on costs, with 2027 expen...
Investor pushback is primarily focused on costs, with 2027 expenses projected to be in the mid-single digits above Street expectations and capital expenditures exceeding $250 billion due to the Muse ramp.
Traditional EV misses ~$628B of off-balance-sheet obligations at 6/30/26, including ~$278B of contracted leases not yet commenced and ~$350B of non-cancelable commitments, >5.6x its ~$112B of on-balance-sheet debt + leases.
Including these obligations lifts FY27 EV/revenue from ~6.15x to ~8.32x, a ~35% increase, making valuation materially richer than peers.
Daily active users (DAUs) on the Muse platform are approaching 2 million, indicating strong initial adoption.
Direct Muse monetization is expected to grow significantly in late '27/'28 through transaction/acquisition auctions across a large consumer total addressable market.
The company has ample capacity to support a large number of daily active users, with '26 CPU purchases alone capable of supporting approximately 1 billion DAUs.
There is a large enterprise/API opportunity, with subscriptions across the company's businesses and an eventual API business, potentially adding significantly to EPS.
The CEO helped shape voluntary AI safety principles, backing a light-touch approach and opposing calls for government-coordinated slowdowns in model development.
The company sees an opportunity to gain share by remaining open to agentic commerce, contrasting with competitors who are blocking such agents.
Needham argues META’s 0% Muse Connectors fee directly challenges AAPL’s 15– 30% App Store take; if 20% of App Store revenue migrated, it estimates ~$10B revenue and ~$7.5B EBITDA downside for AAPL.
MS also frames enterprise as a ~$25T knowledge-work opportunity, with subscriptions across META’s ~100M businesses and an eventual API business, where just 100MW of dedicated capacity could add ~4%/~$1 to ’28 EPS. After the Muse-driven rerate to ~21x, MS
Morgan Stanley's analysis suggests that while direct monetization of Muse will be minimal until late 2027/2028, CPU purchases for 2026 could support a large number of daily active users, and a relatively small amount of dedicated API capacity could significantly boost 2028 EPS.
Similarweb data indicates that Muse is rapidly approaching a significant number of daily active users.
META +10bps as MS ran the Muse math 23 days in: direct monetization stays minimal until late ‘27/’28 (transaction/acquisition auctions across a ~$30T consumer TAM), ‘26 CPU purchases alone could support ~1B DAUs, and just 100MW of dedicated API capacity cou...
Further upside depends on adoption, commercial behavior, and competitive response after the Muse-driven re-rate.
Muse is still failing many real-world shopping tasks, indicating limitations in its current capabilities.
The stock was the only mega-cap to decline, suggesting it was left out of the broader market rally.
The stock continued to digest recent strength, indicating a period of consolidation or pullback.
DASH -1.9% launched a text-to-order agent inside Apple Messages at its DashForward event and said it’s in talks with META on Muse. Useful stat for the agentic-commerce debate from Citizens’ notes with the CMO: early agent traffic cut basket size by two-thirds.
The new Enterprise Platform will package Muse Agent, Meta Business Agent, Muse API, and Muse Code for enterprise customers, creating a potentially more durable, less macro-sensitive revenue stream and clearer ROI on AI infrastructure spend.
Meta’s consumer data, business relationships, and communications footprint are differentiators.
AVGO: Hearing 3p saying TSM CoWoS allocation trimmed to mid 200s from high 200s due to TPU 8i/SunFish ramp pushed out and uncertainty on META volumes.
The company's stock bounced back after previous declines, partly due to OpenAI ceding the consumer agent space to META.
The company is expanding Muse into small business with new integrations and launched its Forum app for Facebook Groups in the US.
BofA and Citi were positive on the new Enterprise Platform.
Second, Dev Day. Less ‘AGI has arrived,’ more ‘please contact your enterprise sales rep.’ The headliner was Dots: always-on agents with their own cloud computer and browser and 4,000+ app connectors, pitched squarely at work and bundled with paid plans (plu...
Enterprise go-to-market (GTM) and product capabilities remain unproven and will require continued investment.
AVGO +1.6% shrugged off 3p chatter that TSM trimmed its CoWoS allocation to the mid-200s from the high-200s (TPU 8i/SunFish ramp pushed out, META volume uncertainty); the S-1’s $161.2B of AVGO-related leases is the more important number I heard investors fo...
META +3.3% bounced off Monday’s -5%, which was partly about fears of an OpenAI Muse rival at Dev Day although we thought it was just also sensible digestion after stock got <15% to the $900 25x target investors generally circling. Dots is pitched at work an...
NVDA +1.7% after adding $150B to its buyback (to $235B, the largest-ever US authorization), launching an Open Agent Safety Platform on a day AI-agent breaches were in the headlines, and The Information reported China is weighing approvals for BABA/ByteDance...
NVDA was the standout +1.7% in semis after announced $150B buyback on top of their existing authorization. Software -55bps held in far better than the headline of the day would suggest (META going after enterprise AI with MDB’s -18% CEO) as cyber ripped and...
DASH -6%: didn’t hear much here other than some chatter being passed around of fears CEO Tony Xu leaving to go to META.
DDOG +1.5% as Wolfe says checks show increasing likelihood of being involved with META’s Muse.
The company is positioning Muse as the core of its consumer AI platform, expanding its capabilities with real-time voice/avatar, Mac computer control, its own email address, and hands-free access through AI glasses.
The Muse connector ecosystem has expanded significantly across various e-commerce and productivity apps, with Shop Pay/PayPal enabling transactions, and the company expects to monetize Muse by taking a small fee on these transactions.
The company broadened its hardware distribution with new Ray-Ban Meta Audio glasses, Gen 3 glasses, over 100 AI-glasses configurations by year-end, and new Meta VR Glasses for spring 2027.
The company's backbone build for AI bandwidth growth, moving towards thousands of fiber pairs in long-haul infrastructure, is a strong positive for optical companies like CIEN, LITE, and FN.
BNP Paribas says Nebius’ outlook has improved meaningfully, with a path to ~$22B+ ARR exiting 2027, ~30% above consensus, as META/MSFT capacity ramps and incremental capacity reprices higher.
Wolfe says meetings at ECOC reinforced its positive view on optical, with CIEN, LITE and FN its favorite names. Most notably, META said AI bandwidth growth is forcing a major backbone build, with long-haul infrastructure moving from ~48 fiber pairs toward t...
The company continues its strong performance, reaching 52-week highs and approaching all-time highs, driven by reaccelerating ad spend growth on Instagram/Reels.
The market's perception of the company has significantly improved, shifting from skepticism about AI spending returns to anticipation of its business model expansion.
BNP upgraded NBIS +7% as they see 2027 as key catalyst as scarcity pricing drives materially higher AI Cloud returns, calling out a path to $22B+ ARR exiting 2027, 30% above street as META/MSFT capacity ramps and incremental capacity reprices higher. Rememb...
Morgan Stanley believes Muse partnerships, such as with EXPE and CART, indicate that agentic commerce is moving from discovery towards transaction capture, potentially opening paths to transaction economics as agents increasingly control the customer interface.
Citrini argues that Muse is the first consumer agent to achieve real mass-market product fit, making agentic disintermediation less theoretical and favoring META due to its infrastructure and transaction rails.
Cantor raised its price target to $860, seeing Muse as a potential third AI S-curve with a total addressable market exceeding $1 trillion, positioning META well through its model, harness, UX, distribution, and massive compute footprint.
Cantor expects the cost per task for Muse to fall by over 80% within 18-24 months and believes META can support rapid adoption using only 5-10% of its projected 14GW FY27 compute footprint, despite current heavily subsidized inference costs.
Muse adoption faces access barriers, as Amazon has blocked its shopping access, and human-support concerns, with an Oppenheimer survey showing only 8% of consumers would trust Meta with passwords compared to 30% for Google.
Meta rolled back an internal trial allowing human contractors to handle Muse calls due to privacy and disclosure concerns, despite humans achieving 95-98% call success in internal tests, which was higher than the undisclosed AI rate.
MS says EXPE becoming Muse’s first major OTA partner and CART integrating grocery checkout show META pushing Muse further down the commerce funnel, potentially opening a path to transaction economics as agents increasingly control the customer interface. Fo...
Citrini argues Muse is the first consumer agent to achieve real mass-market product fit, making agentic disintermediation less theoretical as consumers outsource shopping, negotiating, cancelling and switching decisions. Its framework favors META, SHOP, TWL...
Mizuho reiterated an Outperform rating with a $750 price target, stating that Muse's rapid adoption and Connect increasingly position Meta as a consumer AI platform.
Muse's rapid adoption, with ~730K downloads in its first five days and ~2.5M cumulative installs by September 21, strengthens the case for Meta building a broader consumer AI platform.
Connect is expected to showcase Muse-on-glasses, where hardware, model, agent, and commerce converge, potentially creating a new distribution layer for search, commerce, and digital services.
Third-party connectors and transaction-based monetization are seen as meaningful longer-term upside opportunities.
Tae Kim believes Muse's #1 App Store ranking and Meta's 11% stock move indicate investors are beginning to price personal AI assistants as a major new consumer platform.
Morgan Stanley estimates that every 100 million Muse users could add over $1 billion of revenue and approximately $0.35 of 2028 EPS.
StoneX maintains a Hold rating, arguing that strong adoption of Muse adds to, rather than resolves, Meta's AI return on invested capital debate.
Muse's product-market fit is validated by its ~730K U.S. downloads in its first 10 days and #1 App Store ranking, but it materially increases infrastructure intensity.
Subscriptions are primarily expected to cover compute costs rather than generate meaningful standalone profit, with longer-term upside more likely from commerce referrals/transaction economics and reinforcing WhatsApp/Instagram/AI Glasses engagement.
Agent quality and competitive response remain key risks for Meta's consumer AI platform.
Tae Kim believes the ultimate app winner in personal AI assistants is highly uncertain given coming competition from OpenAI, Google, Apple, and others.
MSFT +1% leading Mega caps higher while META -70bps digesting yesterday’s big move (2nd biggest non EPS ever). Here’s META fwd returns after big moves (h/t J. Fazzuza at Jefferies for the idea):
Muse reaching #1 in the U.S. App Store with ~730K installs in its first five days, significantly outpacing ChatGPT, suggests it could be a meaningful consumer AI winner for the company.
Muse has the potential to drive new engagement and monetization through ads, subscriptions, and transaction revenue, while also validating the company's substantial annual AI investment.
Successful scaling of Muse could support a rerating of the stock towards 25x 2027 EPS, implying a stock price of approximately $810, with a 30x case suggesting approximately $975.
Sensor Tower data indicates Muse is setting new U.S. download records and achieving materially faster DAU growth than ChatGPT's early ramp, suggesting a new product cycle.
The company's 3.6 billion user distribution and over 10 million business ecosystem provide significant advantages for Muse, and its architecture with dedicated secure VMs and a supervisory agent offers potential differentiation in trust and safety.
Muse's success reinforces the shift towards mainstream agentic applications, which is expected to increase server CPU demand for inference, orchestration, and infrastructure workloads.
Muse is expanding its distribution with a Mac launch and TV ads, broadening its reach.
Mizuho says checks remain constructive on AMZN Trainium3, SpaceX’s targeted 7– 10GW 2027 ramp and continued MSFT/META/ORCL deployments, while 1.6T AEC adoption is still early and could surprise higher next year. Mizuho also sees a longer GB300 tail into 202...
The company's stock experienced a significant rally, marking its second-largest non-EPS related move, as investors connected its consumer agentic harnesses like MUSE with future compute demand.
The potential for MUSE to significantly boost the company's advertising business was highlighted by a Microsoft employee's tweet.
ARM +17% the standout winner today, playing catch up to AMD after CEO up ticked last week. Some more color from our weekly yesterday: “If OpenAI announces their own consumer agentic product, we think there might be some slight near-term “narrative” shift in...
The stock continues to exhibit strength ahead of META Connect, with bulls hoping for a near-state-of-the-art Watermelon model.
Bulls are hoping the stock can finally break through the $690/$700 overhead resistance that has persisted since last year.
The stock was weak early in the day due to chatter that OpenAI moved back its "something exciting" launch to coincide with META Connect, potentially a Muse competitor.
The CEO argues that AI safety is best driven by each lab's own incentives and responsibility, with alignment becoming a competitive advantage and most compute focused on serving users.
The market understands that pacing, safety, and alignment in AI will require more compute and spending.
The company's Connect event should clarify its AI product roadmap, with a focus on Muse updates, model progress, Meta AI adoption, and next-gen glasses, leading to an upside 90-day Catalyst Watch and a Buy rating.
Muse downloads are accelerating organically, and there is potential for monetization across transactions, subscriptions, and ads.
The company's model cadence has compressed significantly, indicating faster development.
Meta One pricing creates meaningful subscription optionality, with every 1% consumer penetration estimated to be worth ~$2.5B of annualized revenue and every 1% enterprise penetration worth roughly ~$1B.
Business Agents, paid developer access to the company's LLMs, and the personal-agent opportunity are seen as the three most credible incremental growth vectors.
The stock continues to outperform competitors since the Muse launch, with the CEO emphasizing continued investment in compute for user services rather than just research.
Citi opened a 90-day catalyst watch into Connect, and RBC sees Meta One as a significant monetization lever, with every 1% consumer penetration potentially generating ~$2.5 billion in annualized revenue.
META +50bps contineus to outperofmr AMZN -1% and GOOGL -60bps since the Muse launch. Zuck was out yesterday in an X post basically saying they aren’t slowing down and most compute focused on serving users rather that RSI. Citi opened a 90day catalyst watch...
The company plans to launch camera-free smart glasses, which could be seen as a response to mounting privacy concerns.
Muse continues to receive good reviews and remains #2 on the Apple app store despite only launching in the U.S.
Bloomberg reported that the company plans to deploy its MTIA 450 inference chip in 1H27, with over 1GW of capacity committed over the following 12 months and Astrid ramping by year-end 2027, with early silicon tracking within 2-3% of simulations.
The company's strategy is increasingly focused on low-cost, power-efficient general-purpose inference, reducing reliance on NVDA at scale while remaining a positive read-through for AVGO/TSMC.
META +75bps as Muse continues to receive good reviews and remains on #2 on the Apple app store despite only launching in the U.S. Bloomberg our reporting META plans to deploy its MTIA 450 inference chip in 1H27, with >1GW of capacity committed over the foll...
Muse could generate ~$30B of subscription revenue at just 5% penetration of the company’s ~3B users paying $20/month, with additional upside from commerce take rates.
The company’s data, 3B+ user distribution, and first-mover advantage are real assets.
Recent selloffs in other companies over agent disintermediation are too severe given their fulfillment, inventory, merchant relationships, subscriptions, and proprietary data moats.
Software +1.5% the main beneficiary of the talk so far on rotation and as a “potentially” slower frontier takes some urgency out of the SaaS-disruption bear case. Cyber +4% strongest sector. Internet +50bps also up led by META/GOOGL +2%
The company is seen as a beneficiary from discussions around slowing the AI frontier, suggesting a favorable position in the evolving AI landscape.
The company's new app, Muse, is experiencing rapid adoption, with daily downloads surpassing those of Threads, WhatsApp, and Facebook, indicating strong consumer interest and effective distribution leverage.
The company is positioned to benefit from AI commoditization, as value would shift to its dominant complements like distribution, attention, personalization, and commerce.
Even if AI does not commoditize, the company is not dependent on competitors and can maintain a competitive edge by staying close to the AI frontier.
The company's strategy involves using AI to reduce user serving costs and increase revenue through improved advertising, leveraging its existing cash flow and infrastructure to undercut competitors.
The company has a massive potential install base of 3.6 billion people for its AI, who are less concerned with being at the absolute frontier of AI development.
The company's leader is characterized by a relentless drive and methodical approach to improvement, which is seen as a significant advantage.
Consumer trust around payment credentials remains a relative weakness compared to Google.
JPM upgraded the stock to Overweight with an $820 price target, believing that Muse and frontier models will open new monetization avenues beyond advertising.
JPM believes Meta's accelerated model cadence and distribution to approximately 4 billion users will create significant upside in consumer agents, enterprise intelligence, and Model API access.
JPM highlights strong early traction for Muse, reaching as high as #3 in the U.S. App Store with usage approximately 10 times that of training cohorts, and anticipates eventual monetization through commissions/take rates and subscriptions.
Piper sees improved monetization for Meta, with CPM growth accelerating to +33% from +29% in July, WhatsApp Business pricing increasing by 15–30% in select markets, and early Muse usage running approximately 10 times internal testing levels.
META: JPM upgrades to OW/$820, sees Muse + frontier models opening a new monetization leg beyond ads
META/RDDT: Piper sees stronger August ad trends, accelerating META monetization and solid Reddit user growth
Piper notes RDDT users rose 8% m/m in August and audience growth accelerated to 18% y/y, although a methodology change should drive a ~20% reported reset in audience size. Piper also sees META monetization improving, with CPM growth accelerating to +33% fro...
JPM sees META’s Muse as a formidable consumer agent, but argues SHOP should benefit as agentic channels drive higher-intent traffic, with new-buyer orders from AI sources running at ~2x traditional channels in 2Q.
Piper says ARM is benefiting from hyperscalers increasingly adopting in-house Arm-based server CPUs, with further upside from new custom silicon wins such as META. Piper also points to work with SoftBank-affiliated Graphcore and Ampere on a GPU-like acceler...
The company is well-positioned in the AI space with a mass-market product that requires no unique intelligence, leveraging its existing infrastructure and distribution.
Oppenheimer believes Muse is unlikely to materially impact EPS, citing trust and willingness-to-pay as significant hurdles.
Oppenheimer's survey indicates that only 19% of consumers would pay for an AI agent and only 8% would trust Meta with passwords, which limits Muse's utility for commerce and task execution.
Oppenheimer estimates that a 20% EPS uplift would require approximately $27.5B of agent revenue at 80% incremental margins, equivalent to about 115 million subscribers paying $20/month, which is considered a high bar.
Oppenheimer views improved Siri/Apple Intelligence via Gemini as a competitive risk to Meta's AI offerings.
Analysts are skeptical about the revenue potential of AI agents, viewing the subscriber targets required for significant EPS uplift as a high bar.
META -1.4% giving some back after the big Muse rally yesterday despite the JPM upgrade. Oppenheimer was a bit more mixed on the launch saying a 20% EPS uplift would require ~$27.5B of agent revenue at 80% incremental margins, equivalent to ~115M subscribers...
The Muse app is gaining significant traction, with early usage exceeding projections and users engaging 10 times more than testing cohorts, indicating strong product-market fit.
The company's distribution advantage allows for accelerated adoption of successful products like Muse, which is expected to move both ends of the valuation range higher.
The emerging consumer agent opportunity, exemplified by Muse, is estimated to be worth at least a couple of multiple turns, potentially moving the risk case towards $600.
Muse could enable the company to participate in transaction fees in addition to ad revenue, effectively allowing them to "double dip" on both ad and transaction fees.
The rollout of Muse has been impressive, and the initial +4% stock gain is expected to hold.
The company possesses significant compute infrastructure, including gigawatts of capacity and decades of data center experience, positioning it as a leader in this area.
The company is on the verge of shipping "Watermelon," a significantly advanced pre-trained model, which represents a major advance.
The business model for agents is expected to involve taking a small cut from businesses, as the agents are designed to generate and save money for users, making them self-paying.
The company's distribution and proprietary consumer data across its platforms provide a meaningful advantage in personalized agents, especially if integrated into Muse.
There is little Muse upside currently embedded in the stock's valuation, suggesting potential for rerating with successful execution.
META: MS sees Muse opening a ~$30T consumer-agent TAM, but says adoption/monetization must prove out before estimates move
We’re also hearing plenty of frustration from GOOGL -2% investors asking why Google wasn’t out in front of this, with rotation into META potentially adding to the pressure today. This release should only intensify the agentic fears that Instinct has already...
The stock saw a slight decline following the release of META Muse Code, a personal AI agent.
The ~$18 billion settlement for teen-harm litigation resolves a major overhang, with teens contributing a small percentage of revenue and average usage already below the new daily cap, limiting direct financial impact.
The stock outperformed other hyperscalers, with bulls pointing to an attractive risk/reward profile and upcoming catalysts.
The primary uncertainty is whether behavioral restrictions will reduce engagement among heavy teen users or shift market share to competitors like YouTube and TikTok, especially if peers are not subject to similar limits.
Morgan Stanley believes the youth-engagement settlement is manageable, as teen revenue constitutes only about 1% of total revenue and the 2-hour daily limit is less restrictive for Meta compared to YouTube.
Morgan Stanley draws a parallel to Google's 2025 DOJ resolution, after which product and model launches drove a significant rerating, suggesting a similar setup for Meta with upcoming product launches like Hatch, Watermelon, and Connect.
Bank of America notes that Meta could launch Hatch in early September within Instagram/WhatsApp, providing a significant distribution advantage for the AI agent.
Positive early reviews for Hatch could materially improve sentiment on Meta's AI spending by adding subscription revenue and richer first-party data for content and ad targeting.
Bank of America highlights that the key debate for Hatch is whether it is sufficiently differentiated from competitors like OpenAI/Gemini and trusted enough for autonomous actions, which could lead to slower adoption compared to chatbots.
The stock was weak, potentially due to OpenAI's announcement of its new ad model reaching $1 billion in ARR, which could pose a competitive threat to Meta's ad-centric business model.
SHOP +2.7%: Some color on META/SHOP from Ummera in Slack
The company is discussing a potential mid-trial settlement in a teen social media addiction case, which could involve significant financial liability and mandatory changes to product design.
The company is targeting a late-August/early-September launch for its OpenClaw-inspired Hatch AI agent, with potential premium pricing of up to $199.99/month, and its next flagship model Watermelon is targeted for October.
Meta could begin rolling out third-party AI agents inside WhatsApp as soon as this week, accelerating efforts to monetize AI and compete with OpenAI/Anthropic.
The company could generate approximately $11B of annualized revenue for every 0.5GW of AI capacity sold externally, adding roughly $1.44–2.16 of EPS, or ~4–6% versus 2027 consensus, with a potential ~$11–22B annual opportunity if it leans into external compute.
The optionality of external compute adds to the AI infrastructure bull case, supporting a price target increase.
The company is reportedly setting up to launch a consumer AI agent called Hatch, similar to GrokBot and OpenClaw, with its next major AI frontier model, Watermelon, coming out in October.
Hatch is part of the CEO's ambition to monetize the company's significant AI investments and diversify revenue streams beyond advertising.
The company has considered a tiered pricing system for its AI product, including a premium monthly subscription of up to $199.99 for higher usage limits.
The risk/reward is skewed to the upside, as the stock is trading at 16x street numbers and less than 15x buyside's $35-36, even before accounting for potential revenue from selling compute.
There is a potential $11–22 billion annual opportunity if Meta leans into selling compute, although Mahaney believes today's premium compute pricing will fade.
Bulls believe META has a durable incremental revenue opportunity of over $15 billion from business agents and subscriptions alone, with additional upside from WhatsApp paid penetration and Cloud/API monetization.
External compute is not viewed as a new neocloud-style business, as today’s premium compute pricing is expected to fade as capacity expands, and management remains focused on monetizing higher-margin "intelligence" rather than selling raw compute.
BofA maintains a Buy rating, viewing META at approximately 16x 2027 GAAP EPS as attractive with upcoming AI catalysts including Connect, Watermelon, proprietary silicon, and enterprise AI launches.
BofA sees the current California bellwether social media trial as headline risk, potentially creating meaningful headline volatility through a large jury award and possible product-design remedies, with uncertainty lasting through year-end.
The jury verdict is advisory, the judge has already dismissed several claims, causation remains a key hurdle, and any damages/remedies could be reduced or appealed, making this just one early data point in a litigation process likely to take years.
Bloomberg reported that Meta has become one of Microsoft’s largest AI customers, spending hundreds of millions annually on Azure Foundry and consuming trillions of tokens weekly, including OpenAI models.
The larger implication is that Microsoft’s AI revenue remains highly concentrated among tech companies, with OpenAI contributing approximately 70% of AI revenue last fiscal year.
Meta is exploring its own API business, which could eventually reduce its reliance on external platforms.
The opening of Meta’s youth-liability bellwether trial presents a significant headline overhang, with 29 state AGs seeking substantial damages and injunctive remedies that could force platform changes.
The larger stock risk is not the damages themselves, but any order affecting behavioral AI models, ranking systems, or minors’ data, which could negatively impact engagement and long-term growth.
Bulls argue that the reported $1.4 trillion legal claim is exaggerated and theatrical, with attorneys general already reducing their demands significantly.
The company faces a 29-state trial alleging that Facebook and Instagram were designed to addict young users, misled consumers on safety, and improperly collected children's data.
Potential penalties from the lawsuit could be in the hundreds of billions, and states are seeking nationwide product changes including age restrictions and an end to infinite scroll.
The stock experienced its second consecutive day of weakness as a 29-state trial commenced.
Bears contend that the legal risk is unquantifiable and a potential multi-billion dollar settlement could strain the capital expenditure budget.
Questions persist regarding the return on investment for the company's capital expenditures.
CPM growth accelerated significantly in July.
Piper says July ad buyer checks were broadly positive, with market growth roughly 130bps above expectations and 3Q spending estimates moving about 50bps higher. Piper says RDDT and ROKU saw the largest positive revisions, while META CPM growth accelerated t...
Mark Zuckerberg detailed plans to release a new open-weight model called Muse Glimmer and an open-weight version of its most advanced model, Muse Spark 1.2, in the coming weeks.
The CEO's statement about being "very close to having substantially stronger models" is a bullish indicator for the company's AI efforts.
Recent releases like Muse Spark and Muse Glimmer show promising performance, with Muse Glimmer outperforming similarly sized models on agentic benchmarks.
Upcoming releases, such as Muse Spark 1.2, are described by Meta as a step "toward the frontier" in coding models, with larger models still to come.
The prevailing view is that Meta's AI efforts are still lagging behind competitors like Anthropic and OpenAI, despite significant investment.
The company's stock price increased after releasing a new local AI model, indicating positive market reception to its AI advancements.
The CEO's letter outlined a plan to resume releasing open-source AI models and deliver personal superintelligence, suggesting a strong commitment to AI innovation and distribution.
The release of Muse Glimmer, a 30B parameter dense model that can run locally, and the upcoming Muse Spark 1.2, demonstrate the company's progress in open-source AI models.
The Muse Glimmer model uses quantization techniques to compress the model's weights to approximately 4-bit precision, allowing it to run within a 24 GB or 32 GB envelope with minimal degradation on agentic tasks, showcasing efficient model deployment.
The stock rose after releasing a local model that rivaled Opus 4.5 performance, indicating strong AI capabilities, and the CEO wrote a long AI letter.
The company debuted an AI Coding Agent.
Meta is still growing faster than any other at-scale ad platform, indicating strong performance in its core business.
The stock outperformed, benefiting from the positive sentiment around hyperscalers and the broader ecosystem's positive return on investment.
Just like GOOGL (+5%) played catch up to MSFT/AMZN (+5%) on Friday, today it was META’s (+6%) turn to outperform, along with the 2nd order trades: ORCL (+9%) and the Neoclouds (CRWV +19% / NBIS +11%), the rationale being that positive ROI is good across the...
The company is currently double-paying for infrastructure by both building its own data centers and renting compute from third parties, without a clear path to monetization for these investments.
Meta faces a timing mismatch where it is making significant investments in AI infrastructure now for future benefits, which investors find confusing given the lack of immediate monetization.
The company's "improving monetization" story has lost some luster, and investors question why Meta isn't solely focusing on its profitable ad business.
Zuckerberg's focus on addressing Meta's existential risk in the face of AI means the company is trying to sell investors on a future vision while making investments that frontier labs made in the past, creating a second timing problem.
RDDT is competing for budgets with much larger platforms that have stronger identity signals, broader commerce data and more mature automation. Recent reporting also points to growing competition from META’s Threads and forum-like products. The licensing di...
The risk/reward profile for the stock remains skewed to the upside, a long-standing view, and the stock is considered a long-term holding.
AI spending is already yielding measurable returns within the core business, evidenced by strong revenue growth, advertising growth, increased Instagram time spent, and improved Facebook video time and ad conversions.
The near-term revenue slowdown is primarily due to tougher comparisons and not indicative of losing ad share or weakening advertiser demand.
Street estimates do not fully account for the potential contributions from new revenue streams like Business Agents, Meta One, model APIs, and potential compute sales.
The company's extensive user base and advertiser relationships provide a cost-effective path to scale consumer and business agents compared to standalone AI startups.
The stock's valuation at 15x CY27 numbers suggests limited downside risk.
The core advertising strength was already factored into expectations and did not address concerns about incremental earnings and cash flow from significant infrastructure investments.
The Q3 midpoint missed expectations, no direct compute deal was announced, personal agents are not yet launched, and management did not provide a framework for 2027 capex.
Management's comments about maximizing 2026-27 capacity and preserving growth in 2028 could indicate an open-ended spending commitment rather than a capex peak.
The business is becoming increasingly fixed-cost heavy, with adjusted operating margin compression despite strong revenue growth, rising stock compensation, and declining free cash flow.
The enterprise effort requires new go-to-market capabilities, and personal agents, coding tools, and APIs place the company in direct competition with established players.
Legal risk adds uncertainty, highlighted by a significant charge and warnings of potential youth-related losses.
The company delivered a disappointing quarter with top-line revenue falling short of buyside expectations.
A modest increase in operating expenses was viewed negatively, contrary to expectations for a flat or slightly lower trend.
The capital expenditure raise was limited, with the midpoint only increasing by approximately 2% and the high end remaining unchanged, which disappointed buyside and AI semi bulls who desired a larger raise.
Operating income appeared to be a slight miss, although it was a small beat when excluding one-time fines and severance costs.
Ad impressions were down compared to the previous quarter.
FundaAI’s agency checks suggested Meta’s 2Q advertising performance was broadly above expectations, with Reels, Overlay Ads, and Andromeda-driven recommendation improvements boosting inventory, conversion, and ROAS.
FundaAI identified Business Agent as the strongest product datapoint, with agencies viewing it as a potentially significant monetization opportunity spanning subscriptions, payments, messaging, and automated campaign management.
SHOP +12% as Cleveland was out positive saying Q2 finished well ahead on GMV/payments upside and share gains. Sentiment here still remains mixed - buyside wants to get more positive on some sort of price increase in 1H’27 and likely upside to top line while...
The stock showed weakness alongside AMZN.
Hypers mixed: GOOGL +2% getting some love after last week’s disaster but META/AMZN -20bps.
Other hyperscalers underperformed: META +66bps; AMZN flat
The stock underperformed compared to other hyperscalers.
Neoclouds weaker as investors digest the META “compute” news from last week: NBIS -8%; CRWV -4%. Charts not looking so hot for either as NBIS now down
Online ad spend came in ahead of expectations in Q2, helped by a stronger June, World Cup activity, and Amazon's shifted Prime Day timing, with performance budgets outpacing brand spending.
The company is a clear standout in sponsored ads, DSP, and CTV.
Stronger ad pricing and emerging monetization from excess compute, model access, and Business Agents support EPS and multiple upside.
Muse Spark 1.1 materially improves multimodal capabilities and appears competitive with leading models, potentially reducing the company's prior AI model disadvantage.
External monetization through APIs, business agents, and raw compute, while a potential upside, remains unproven in terms of pricing, partner traction, and enterprise adoption.
JPM wants clearer evidence that Muse can drive durable external revenue without materially higher capex or renewed returns concerns.
The stock was digesting its significant move from the previous week.
META -2% digesting last week’s big move while GOOGL -1.3% continues to roll over.
The company views its API business for selling access to its AI model as a significant revenue opportunity, with tokens growing rapidly, suggesting it can become a meaningful business even at Meta's scale.
The company is optimistic about its ability to be at the forefront of AI by the end of the year.
The company's target of 14GW of compute capacity additions across 2026–27, implying a significantly lower cost per GW than estimated, suggests engineered cost savings and positive economics compared to competitors.
The Iris custom chip is seen as a positive for sentiment, supporting the return on investment for capacity.
The company's AI product cycle, including strong benchmark reception for Muse Image, business-task performance of Spark 1.1, and API pricing competition, is expected to drive positive revenue revisions and P/E expansion.
The company is expected to deliver in-line revenue for Q2 and upside to typical Q3 seasonality due to AI products and wearables.
Checks indicate that the Family of Apps continues to generate healthy ROI despite macro and geopolitical ad-budget risks.
The leaked memo indicating 14GW by 2027 and capex up to $145B suggests an escalation of ambition without providing clarity on the return on invested capital, which is a key concern.
The setup for the second half of 2026 and 2027 skews towards downside risk without explicit unit economics or performance-per-watt disclosure.
The company plans to start manufacturing an artificial intelligence chip from September as part of its plan to boost overall computing power to 14 gigawatts next year, indicating a significant investment in AI infrastructure.
The relatively quick progress in testing its in-house AI chip, with no major issues found in six weeks, signals positive momentum for an effort that had previously struggled.
The company was added to BofA's 'US 1 List'.
The stock was down due to the implication of higher capital expenditure numbers.
The company is testing 'super sensing' AI glasses that can capture every moment, pushing into the contentious market for all-seeing, all-hearing devices.
The stock performed weakly, indicating a negative sentiment among hyperscale companies.
Hypers weaker: META -2%, AMZN -1%, GOOGL
Meta's massive AI investments should be viewed as a bet on strengthening its core advertising business, not reinventing it.
AI will dramatically improve content recommendations, ad targeting, and monetizable inventory.
The compute buildout is essential to defending its digital franchise and driving a new phase of long-term ad growth.
Investor concerns around Meta potentially renting excess AI capacity through a cloud offering are overdone and instead reinforce the long-term bull case for power infrastructure.
Selling excess compute could generate attractive returns and help Meta monetize temporary surplus capacity before internal AI products fully scale.
Meta ad spend rose low double-digits as pricing accelerated despite softer impressions.
Meta Business Agent is emerging as a potentially meaningful new monetization lever if it expands WhatsApp and SMB engagement.
The company's stock rose after announcing the release of its first AI image creation model, Muse Image, which will be available to consumers and power users.
Four states are seeking $1.4 trillion in penalties over accusations that the company designed its platforms to addict young users and misled the public about their safety.
Meta’s superintelligence chief Alexandr Wang stated that the company’s upcoming AI model, codenamed Watermelon, has caught up with OpenAI’s flagship GPT-5.5 model, based on closely followed AI model benchmarks.
SemiAnalysis concludes that Meta still aims to be a leading frontier AI model provider.
Any move to sell excess capacity to third-party customers will be differentiated from plain vanilla IaaS providers and highly accretive to earnings/cash flow.
Wang suggested that the company's models have caught up to Chat GPT 5.5, and Semianalysis defended its strategy to sell compute.
The company's positioning in the AI world is strong, with Llama being a mainstream U.S. open-source model, which carries a national security premium similar to Intel.
The company is buying optionality by developing a system to rent out excess compute between training batches and reallocating H100s for external inference, potentially building a cloud business.
GOOGL +2% / META +3% leading hypers higher while AMZN +60bps lagged. Wang was out saying METEA’s models caught up to Chat GPT 5.5 while Semianalysis also defended the move to sell compute. We flagged this in our morning wrap on META from Slack, which w...
Needham argues Meta’s AI investment strategy is leading to a deterioration in returns on capital, citing rising capital intensity, growing stock-based compensation, and expansion into multiple capital-intensive businesses beyond its core advertising franchise.
The firm believes Meta has overbuilt AI infrastructure and questions the economics of entering cloud services.
Needham warns that higher CapEx commitments and strategy diversification could weigh on ROIC for years.
While acknowledging Meta’s ability to monetize excess AI capacity, Needham remains skeptical that the investments will generate sufficient incremental returns to justify the current pace of spending.
SemiAnalysis concludes that the company will continue spending gargantuan amounts of money on capex, with capex in 2027 expected to be shockingly high.
Meta's large-scale deployment of its custom Vistara CXL ASIC validates CXL as a production-ready technology, increasing investor confidence in the long-term CXL ecosystem.
The deployment highlighted reported 20-25% compute capacity savings and 4-12% throughput gains, reinforcing the long-term investment case for CXL across datacenter memory and server infrastructure.
The company's aggressive buildout of owned AI infrastructure positions it to potentially monetize excess compute capacity, which could generate significant revenue and EPS, making its CapEx returns materially better than market assumptions.
The company benefits from durable measurement and attribution advantages in advertising, particularly with GenAI budgets being largely incremental.
The stock experienced weakness due to ongoing issues plaguing its narrative.
New consumer and business subscription offerings across its platforms represent a meaningful new monetization opportunity that is largely ignored by investors.
Modest adoption of new subscription products could generate over $10 billion in incremental revenue and approximately $5 billion in operating profit by 2028.
Subscriptions provide a high-margin revenue stream not reflected in Street estimates, potentially adding to its 2028 revenue and operating income forecasts, creating a new growth driver beyond advertising.
There is a larger opportunity for Meta to monetize excess capacity through a 'neocloud' offering, with potential to add approximately 10% to 2028 EPS if 500MW were leased at ~$25/watt.
Modest adoption of subscription products could generate substantial incremental revenue and operating profit by 2028, including from consumer app subscriptions, Meta AI premium tiers, and business-focused Meta One offerings.
Meta's CTO stated that morale at the company is near an all-time low due to recent mass layoffs and internal AI initiatives.
The CTO indicated that employee morale is near its lowest point.
The head of product for "AI for work" transformation is leaving the company.
The stock generally outperformed, consistent with large-cap trends.
Large cap generally outperformed with GOOGL/META +1% and AMZN flat.
AI-referred traffic appears to have inflected higher in 2026, with Facebook seeing measurable increases, albeit still generally below 1% of total traffic.
The stock saw a tactical bounce after one of the worst weeks for mega-cap stocks versus the market in recent years, suggesting a potential rebound.
Tensions within Meta's Applied AI organization are escalating, with employees describing the work as repetitive and demoralizing, and morale pressured by restructuring, layoffs, and increased monitoring.
The stock showed early strength, gaining 1-2%, alongside other hyperscalers.
In Tech, some hyperscalers strength early with AMZN, META and GOOGL +1-2%. Some strength in ALAB, NBIS, CRWV (all +5%) as they get added to Nasdaq 100.
Memory/HDDs: Hearing Semianalysis positive on NAND pricing calling out 50c/GB for META/GOOGL LTAs and that HDDs could see 30-40% yy px hikes in ‘27
BofA sees clear near-term benefits from AI through improved ad targeting, recommendation systems, engagement, and advertiser tools.
Newer monetization vectors such as subscriptions, business agents, and enterprise AI are beginning to emerge.
Clearer evidence of monetization over the next several quarters could support a meaningful re-rating from current valuation levels.
Edgewater reports mixed ad checks, with more even performance compared to Q1 due to challenged return on ad spend (ROAS), though revenues are tracking in line.
The market is still looking for proof that Meta’s frontier-model investments can generate returns commensurate with rising capex and opex.
Cleveland noted a deceleration in advertising and macro headwinds in Q2 that were greater than Q1, resulting in revenues only being in line.
An Amazon cloud executive asked employees to recruit recently laid-off Meta workers, indicating that Meta has experienced recent job cuts.
Megacap largely weaker: GOOGL/AMZN/META all down more than 2%.
Channel checks indicate positive retail/e-commerce demand, with direct response (DR) being the biggest driver due to strong ROI.
The market continues to underappreciate Meta’s emerging subscription opportunity, which could build into a $20B+ high-margin business over time through products like Meta AI, Meta One, and premium creator/business offerings, diversifying revenue beyond advertising. Management’s strategy is to layer paid features on top of Meta AI’s rapidly expanding user base while using subscriptions to fund continued AI investment. Meta AI’s scale advantage and growing engagement create a credible path to meaningful subscription revenue, potentially making Meta the second-largest advertising platform and a sizable subscription platform simultaneously.
The stock received a positive mention from Cleveland, indicating that the quarter is tracking above expectations.
The stock was a large-cap underperformer, with ongoing debate about the implications of likely equity raises for the group.
Large cap a big underperformer today: AMZN -33bps; GOOGL -1.3%; META -1.3% as debate continues around what likely equity raises mean for the group.
Early signs indicate that the company is beginning to monetize its GenAI chatbot efforts through Business Messaging.
The broader rollout of Business Agent, integrating with platforms like Shopify, Zendesk, and Salesforce, enables businesses to automate customer interactions within the company's ecosystem.
New WhatsApp commerce and scheduling features are underappreciated monetization levers, as WhatsApp remains one of the company's most under-monetized assets.
Recent product launches increase the likelihood that the company's AI investments will contribute to revenue growth in 2027.
Morgan Stanley sees four AI-driven paths to meaningful earnings upside, reiterating a top pick rating.
The largest opportunity is in AI search, with Meta AI potentially generating over $10 billion in annual "search-like" revenue and roughly $3/share of EPS upside by leveraging Meta’s distribution, engagement, and commerce ecosystem.
Subscription offerings across businesses, creators, and consumers represent a potential $7 billion-plus revenue opportunity.
Every 1% upside to 2028 ad revenue equates to roughly $3.5 billion of incremental revenue and approximately $1 of EPS.
Leasing excess compute capacity in a neocloud-style model could provide another meaningful source of earnings upside if Meta’s AI buildout outpaces internal demand.
The company has produced Muse Spark, its most credible AI model yet, released from Alexandr Wang’s secretive TBD Lab.
Critics inside the company describe Alexandr Wang’s leadership as frenetic and argue he has overplayed what is more incremental progress in AI.
Muse Spark was built partly using Llama 4 code and datasets, despite Wang’s suggestion it was made “from scratch,” which irritated some staff.
While Muse Spark has been praised for visual understanding, Wang concedes it trails rivals in coding, with employees testing it still preferring Anthropic’s Claude.
The stock outperformed other hyperscalers, including AMZN.
AMZN -2% as META -50bps outperformed the other hyperscalers
RBC argues the market is underestimating the company’s opportunity to create and monetize new small businesses through AI tools, framing it as a potential sixth growth pillar not reflected in current valuation.
The firm notes the company already reaches roughly 250M businesses across its platforms, with only a small percentage spending more than $20k annually on advertising, creating a large monetization runway.
RBC estimates ~10M new SMBs are formed globally each year, representing a sizeable pool of future advertising demand, while early AI products such as Andromeda are already showing measurable improvements in ad performance.
The analyst believes the company’s AI stack could lower the barriers to business creation and eventually drive incremental advertiser growth.
The company plans to start testing an AI pendant in the next year as part of an ambitious roadmap for wearable devices aimed at reversing the huge losses in its hardware division.
The roadmap also lays out plans to significantly expand its selection of AI glasses and to add a business-focused service called “Wearables for Work,” indicating a strong commitment to AI hardware.
The strategy is in part to drive more use of the company’s AI models and products such as subscription versions of its apps and a consumer AI agent it is developing called Hatch.
The company ranks #2 in citation share for AI-assisted search, reinforcing its growing importance as source material for AI search results.
Adoption and monetization of the company’s AI stack remain in the early stages.
The stock underperformed due to a rotation out of hyperscalers, as investors shifted funds to software names without selling AI semi longs.
Sell-side analysts are estimating that the subscription announcements could contribute approximately 4-5% to revenue and over 10% to EBIT/EPS.
The company launched Instagram, Facebook, and WhatsApp subscriptions, offering extra features for a monthly fee.
The company will begin testing more subscription plans, including Meta One Plus ($7.99/mo) and Meta One Premium ($19.99/mo), with the Premium plan offering more capacity for higher compute queries and enhanced video/image generation capabilities.
These subscription launches support the narrative that Meta will monetize its capital expenditures.
Investors are generally looking to go long on the stock and have been awaiting progress on the AI model front.
The subscription news helps, especially for a stock that has been roughly flat since the QQQs (including GOOGL and AMZN) started rallying in early April.
There is not much pushback from investors about wanting to own META near $600, given the risk/reward profile.
APP +10% following through on yesterdays positive Edgewater note (META pushing back no-IDFA bidding) that cleared that catalyst path for the June GA release.
The stock was largely weaker early.
MU +8% getting close to $1k. Lots of your favorite AI semi names up 4-6%. Internet and Software down early, the latter led lower by ZS’s -25% horrid print, which is bringing down other cyber names as well (PANW/CRWD -4%). Interestingly, hyperscalers largely...
APP +7% / U +5% strong on the back of an edgewater note saying META bidding into 3p no longer expected over the next couple months, saying it sounded more like a 2027 event according to their conversations (Oppenheimer attended the same trade show last week...
Recall - META has been the main overhang on APP’s stock in the near term, and while this doesn’t eliminate it completely, it is a positive near-term datapoint that simplifies the catalyst path for the GA e-comm launch in June.
Cost rationalization actions are expected to drive significant EPS upside, while the Family of Apps maintains durable growth.
GOOGL/META: RBC Says OpenAI Ads Rollout Creates Competitive Pressure, Questions Monetization Path
The co-founders of Manus are exploring options to raise funds to unwind Meta's takeover of the Chinese-founded AI operation due to Beijing's demand.
Morgan Stanley raised FY27 EPS estimates by approximately 3% to $34 while maintaining an Overweight rating and a $775 price target, as expected workforce reductions, slower hiring, and disciplined expense management should help offset rising AI capex.
Morgan Stanley estimates a 10% workforce reduction could drive approximately $3.5B of annualized savings, equating to about $1.20 of EPS in FY27.
The firm highlights a potential 'Neocloud backup' option, arguing that excess compute capacity could eventually be monetized externally and add approximately 8% upside to 2028 EPS.
The company released a presentation outlining its strategy to capture search ad dollars using Muse Spark, highlighting that social and discovery platforms are widely used for product information and influence purchase decisions.
The firm highlights raised long-term server CPU TAM forecasts (~$120B by 2030), strong OpenAI/META/NVDA-related demand, and better-than-expected FY26/FY27 outlook supported by supply finally catching up to demand.
The stock received an unloved bid, outperforming other internet peers.
Meta AI returned to double-digit month-over-month growth in April, indicating strong app performance.
META +70bps getting some of the unloved bid today outperforming GOOGL -30bps and AMZN -1%. This chart getting passed around, although it was out yesterday
The stock outperformed other megacap favorites like GOOGL and AMZN.
Competitive concerns from META are expected to loom in the second half of the year, and the recent print will not change bears' conviction in this thesis.
The company is well-positioned to serve the non-enterprise AI consumer market due to its focus on individuals and existing ad network.
The company can dedicate its substantial compute budget to consumer AI, unlike competitors who also serve enterprise clients.
The company has a shot at being the leading consumer chatbot provider, given its compute resources and previous AI efforts.
The company has the luxury of potentially offering its AI services for free and monetizing later, similar to its typical playbook, unlike OpenAI which needs to charge.
The company, like Google, has the advantage of not needing to fund its consumer AI with ads, unlike OpenAI.
The company lacks consumer trust regarding personal data, which is crucial for productive economic agents.
The company's business trajectory remains very strong, despite a perceived slowdown from an 'insanely strong' Q1 to a 'merely strong' Q2.
The company experienced a 'trajectory change' in its ad business after the start of the U.S. war in Iran, impacting sales.
Increased investment in compute and infrastructure for AI necessitates reducing capital allocation to people-oriented areas, potentially leading to layoffs.
Bernstein sees continued outperformance from large platforms like Meta, supported by resilience in ad trends despite softness in certain verticals and international markets.
Spend is increasingly concentrating with scaled players like Meta, and AI-driven opex could support revenue upside.
The stock is seen as a play on execution and duration, with the expectation that healthy top-line growth, controlled expenses, and occasional investor incentives will drive the stock higher, barring macro or regulatory shocks.
Strong revenue growth is crucial, with expectations for Q1 revenue and a Q2 guide that provides latitude for AI investments and buys time for investors.
The Muse Spark model demonstrates that the company is actively pursuing AI initiatives, providing proof of concept.
China's National Development and Reform Commission ordered Meta Platforms to unwind its $2.5 billion acquisition of AI startup Manus on national security grounds.
There is no clear indication or demonstration of the monetization potential of its AI capabilities, and the timeline or form factor for monetization remains uncertain.
The company needs to introduce new consumer-focused products in images and video, especially since OpenAI has conceded the consumer market.
Jefferies notes that Meta shows stable demand and some CPM/reach improvements in the mobile ad market.
Jefferies notes that Meta shows weaker performance trends despite some CPM/reach improvements in the mobile ad market.
The firm notes U is improving with better Vector performance and higher ROAS (+14%), though still trailing APP, while META shows mixed results with stable demand but weaker performance trends despite some CPM/reach improvements.
WhatsApp is selling a new premium subscription that includes more custom features, part of a broader effort to build a revenue stream that can complement its core advertising business.
Digital ad checks point to resilient 1Q budgets with limited downside risk, and Meta is viewed as broadly solid.
The firm prefers GOOGL into the print given upside potential in Search and accelerating Cloud, while viewing META as broadly solid but with capex scrutiny persisting, TTD as facing near-term noise but supported by easing comps and event-driven tailwinds, an...
Capex scrutiny persists.
GOOGL/META/TTD/PINS: Stifel Sees Solid 1Q Ads but Cautious 2Q Guides; Prefers GOOGL into Print
Meta's initiative to capture employee mouse movements, clicks, and keystrokes for AI training data raises concerns about employee privacy and potential internal resistance.
62 EPS) and continued momentum from cloud/AI customers (MSFT, META, GOOGL); the firm highlights robust backlog, expanding scale-out/scale-up deployments, and incremental tailwinds from upcoming XPU/optical transitions, supporting sustained growth and margin...
Piper says APP continues to see strong gaming trends and improving supply/demand dynamics, with sellers. json data showing a re-acceleration in net adds in April after a softer Q1, while SDK share remains stable (~25% in 1Q26); the firm highlights E-commerc...
The company is targeting May 20 for the first wave of layoffs, with additional cuts expected later in 2026, impacting approximately 10% of its global workforce.
For AMD, the firm raises server assumptions (~50% YoY growth) and incorporates a new META AI deal, seeing upside to ’27 estimates not yet fully reflected in consensus, while noting continued PC softness near term.
Wolfe initiates NBIS at Peer Perform, citing strong hyperscaler demand (MSFT/META commitments, NVDA backing) and improving GPU pricing as validating the demand story, but flags key open questions around execution of large-scale capacity builds and the durab...
AVGO: Bernstein Reiterates Outperform, $525 PT; META ASIC Partnership Expands Long-Term Opportunity
Stuff like this has been floating around for a few weeks now so not new (Rubin shipments just get replaced with more Blackwell): Blackwell @xft) We've heard more positive reports in the supply chain as well; for example; Cleveland was out this morning sayin...
Barclays sees the company's Muse Spark release as proof of its progress from its Superintelligence Lab, putting it back in the AI conversation and potentially driving a narrative shift towards it as an "AI winner" if execution continues.
Barclays added that the company's distribution, data, and compute advantage position it well for consumer AI scale, with products like Manus and Dreamer expanding capabilities, leading to improving confidence in its AI trajectory despite near-term capex pressure.
The debut of Muse Spark, Meta's first homegrown model, is a meaningful catalyst, as benchmark gaps against frontier peers are narrower than feared, and the company's ability to productize first-party model capabilities across its Family of Apps is crucial.
Muse Spark's built-in shopping assistant sets the stage for agentic commerce, with Meta's user/influencer content corpus potentially differentiating it in building trust, a key factor for adoption.
An API monetization bull case for Muse Spark, which is currently closed-source and offered to select partners, adds optionality not reflected in its current valuation.
Morgan Stanley maintains Meta as its top pick, seeing significant upside potential.
CRWV/META: CoreWeave Expands AI Cloud Infrastructure Deal With Meta for $21 Billion
META remains MS’s top pick with ~25% upside.
The company received positive news regarding its model development, announcing the earlier-than-expected release of its new closed model, Muse Spark, which performs well against competitors.
The company's pretraining stack improvements, including model architecture, optimization, and data curation, have significantly increased efficiency, allowing the same capabilities with an order of magnitude less compute than previous models.
The company has the greatest potential for a positive narrative change among large-cap companies due to its large distribution and ability to leverage it for new products, despite starting behind in the state-of-the-art race.
Channel checks indicated a strong finish to the quarter, with positive sentiment for the stock increasing over the past 1-2 weeks.
There were some mgmt re-shuffling yesterday post-close at APP, but move happened because of some mixed channel checks from BTIG/MS (I believe they both talked to Ben Legg), calling out decelerating gaming growth (META taking share) although said Ecom was do...
Ad/GOOGL/META: DB Expert Call — 1Q Spend Beat Expectations, 2Q to Accelerate to on World Cup/USA250, but Visibility Collapsing
The stock experienced a decline in early trading.
Benchmark views Meta's hybrid checkout strategy as defensively motivated against large language model (LLM) disintermediation, raising long-term questions about whether it can make Shopify and merchant storefronts less necessary at the point of conversion.
The company is testing a new premium subscription service for Instagram to diversify revenue beyond advertising and build a more reliable recurring revenue stream.
The company was part of the 'Mega cap strength' trend, indicating strong performance.
Mega cap strength: GOOGL +5%; META +6.5%; AMZN
Morgan Stanley upgraded the stock to a Top Pick, citing that sentiment has bottomed out with shares trading below historical levels and at approximately 15x 2027 EPS.
The firm anticipates durable core advertising growth, exceeding Street expectations, driven by improved engagement and monetization through machine learning.
Morgan Stanley highlighted an emerging opportunity in agentic AI, leveraging Meta's extensive user base, commerce data, and multi-modal platforms to capture spending in agentic shopping and advertiser automation.
Potential workforce reductions could increase 2027 EPS by over $1, thereby raising the earnings floor.
Multiple catalysts, including AI monetization, agent rollout, and cost leverage, are expected to drive earnings revisions and a re-rating of the stock.
MS highlights emerging agentic opportunity (“MetaClaw”) leveraging Meta’s 3.5B+ user base, commerce graph, and multi-modal surfaces, positioning META to capture agentic shopping and advertiser automation spend. The analyst adds potential workforce reduction...
Recent workforce reductions could add $1 to EPS, cushioning against any macro headwinds.
META +2% as MS added to top pick with $775PT, arguing sentiment has troughed with shares at ~15x ’27 EPS and below historical levels.
A Los Angeles jury verdict holding Meta liable for designing addictive products that harmed a young user threatens to expose the company to thousands of similar product-liability lawsuits, potentially forcing fundamental changes to its platforms' features and advertising models.
The verdict, combined with a separate $375 million New Mexico ruling against Meta, signals a shift in legal liability away from content moderation immunity toward platform design practices.
META/GOOGL: Meta, Google Risk Big Tobacco-Like Fallout After Addiction Trial
Digital ad spend is tracking strongest for the company, with pricing remaining firm as advertisers prioritize revenue over margins, supported by sports and election tailwinds.
The firm notes the chip—co-developed with META—targets agentic AI workloads with higher bandwidth and performance, positioning ARM as a more direct participant in AI infrastructure.
Chinese authorities have restricted two co-founders of AI startup Manus from leaving China while reviewing the company's acquisition by Meta Platforms.
RBC stated that META stands to gain from continued digital ad growth.
Goldman Sachs reiterated Buy with an $835 price target, arguing recent headlines around layoffs, Reality Labs cuts, and model delays are likely overstated relative to the company’s broader strategy.
The firm sees potential for cost rationalization to drive upside to EBIT/EPS—ranging from mid-teens in 2026 to 20%+ longer-term—while maintaining investment in AI and compute infrastructure.
Goldman added that META remains well positioned for durable growth across its Family of Apps, with monetization tailwinds from Reels, messaging, and AI-driven ad improvements.
Mark Zuckerberg is building a personal “CEO agent” to streamline his workload, spearheading a Meta-wide initiative to embed AI into every employee’s daily routine, which has fueled internal anxiety that AI-driven efficiency could eventually lead to further layoffs.
GOOGL/META: RBC Says OpenAI Ads Rollout Creates Competitive Pressure, Questions Monetization Path
The quiet shutdown of Horizon Worlds VR, following significant operating losses, serves as a cautionary parallel for AI hype, with limited evidence of AI's positive impact on revenue or costs.
Wells Fargo believes Meta's substantial investment in AI compute capacity by 2028 will position it to lead the consumer AI application layer and support strong advertising performance.
Morgan Stanley notes that Meta is beginning to realize efficiency gains from GenAI tools, with management citing a 30% increase in engineering output since early 2025 due to agentic coding tools.
A potential 20% workforce reduction could generate significant annual savings, providing a buffer into 2027 and supporting continued investment in AI infrastructure.
A potential 20% workforce reduction could generate roughly $3B–$13B in annual savings and provide 5–10% upside to 2027 EBIT.
Meta has committed to purchase additional available compute capacity across certain upcoming Nebius clusters up to a total of $15B over a five-year period.
Meta signed a new long-term AI infrastructure supply agreement with Nebius, committing to $12B of dedicated capacity over five years.
Meta is planning sweeping layoffs that could affect 20% or more of the company as it seeks to offset costly artificial intelligence infrastructure bets.
Meta shares may experience elevated volatility due to the wide range of outcomes tied to its investment cycle.
The CEO has reportedly begun dismantling the power structure of a key executive leading Meta Superintelligence Labs, suggesting internal leadership issues.
The executive's sidelining was evident with the creation of a new applied AI engineering organization under a different veteran executive, and control over key AI models was shifted.
The restructuring is widely perceived as a 'soft firing,' indicating a significant loss of influence for the executive despite retaining his title.
The agreement with AMD for AI chips is estimated to contribute significantly to AMD's revenue, suggesting a substantial investment by Meta in AI infrastructure.
META/AMD: Meta and AMD Agree to AI Chips Deal Worth More Than $100 Billion to Deploy 6GW
The Avocado open-source model has surpassed other open-source models, which is a positive development for Meta.
The company's sustained capital expenditure momentum is directly driving stronger demand for AI servers and ASICs.
The MTIA-2 chip is already in production and slated for a H1 2026 debut, built on TSMC’s 3nm process with advanced CoWoS-S packaging.
The MTIA-3 chip is set for a H2 2026 debut, featuring a more complex design with extra I/O and an additional SoC, also using TSMC’s 3nm process.
The larger reticle of MTIA-3 limits CoWoS packaging to only eight chips per wafer, requiring reliance on GUC’s expertise for execution.
The company is well-positioned to leverage AI for growth through agentic AI for SMEs and consumer entertainment video generation.
META/GOOGL: KeyBanc Lowers PT to $835, Flags Near- Term Product Cycle Lag
KeyBanc lowers its Meta price target to $835 while maintaining an Overweight, the firm says, arguing 2026 product-cycle benefits skew more clearly to GOOGL despite Meta’s valuation already reflecting AI capex and ROI concerns. The analyst notes META trades...
Near-term product cycle benefits are less clear compared to peers, with elevated operating expenses and capex ahead of revenue inflection.
BofA upgraded Oklo to Buy and raised its price target to $127 from $111 following a binding agreement with Meta to develop a 1.2GW advanced nuclear campus.
Meta's willingness to fund development well ahead of delivery underscores the increasing role of nuclear power as AI and data center demand accelerates.
OKLO: BofA Upgrades to Buy on META Deal; PT Raised to
DB frames 2026 around three themes—AI-driven top-line acceleration, easier comps in search and discovery, and a more supportive regulatory backdrop—and names Alphabet, Reddit, AppLovin, and Uber as Best Ideas given asymmetric growth and valuation setups, th...
UBS cut its price target due to a more constrained Q4 earnings setup for ad-driven companies, following a slow October and a narrower beat profile compared to prior quarters.
The company is reportedly scaling back its in-house ASIC plans and instead buying AMD's MI455X.
RBC’s SMB ad checks remain constructive, with consumer softness falling significantly and AI tools unlocking incremental budget as TikTok pressure fades.
The company is seeing outsized auction inflation and stronger AI content tool feedback, narrowing its pricing gap to GOOGL.
Edgew reports a healthy finish to Q4 and encouraging 2026 outlooks.
The company is securing nuclear energy deals to power its Prometheus AI supercluster, indicating a strategic move towards sustainable and scalable energy solutions for its advanced computing needs.
Meta delayed the international rollout of its Ray-Ban Display glasses due to strong demand in the United States and short supply, indicating successful product reception.
Chinese officials are reviewing Meta’s $2bn purchase of AI start-up Manus for possible technology export control violations, which could give Beijing leverage over the transaction.
Futures flat but lots of movement underneath the surface with NVDA -4% and AMD -6%. The sell OpenAI levered names vs Buy GOOGL/TPU exposed names was in full force yesterday and likely continues today after TheInformation calling out META possibly buying TPU...
Other large cap weaker: AMZN -50bps; META -75bps; NFLX -1.6%
AAPL -75nps as another AI lead left for META
Q2 revenues beat both street expectations and the company's guidance range
Q3 revenue guidance is ahead of street expectations
The company raised the low end of its operating expense guidance
Capital expenditures guidance was raised by $2B
Resilient topline growth in Direct Response ads, strong AI execution, and heavy investment in AI infrastructure.
The coming week is packed with reports that should help fill in some blanks in tech as Q3 gets under way. Expect meaningful updates across a wide spectrum: a clearer picture of hyperscaler capex trajectories (META, MSFT, AMZN), critical reads on the pace an...
RDDT +1.3%/ META +7Obps / AMZN +60bps ahead of earnings on Thursday DUOL -6.5% as 3p continues to show a deceleration in paid bookings through the end of Q2.
Elevated AI capex is expected, and there is some compression in ad growth spreads vs. peers.
AI-driven ad performance improvements
Monetization opportunities in Threads and WhatsApp
Research through 2026-11-03 · 2025-06-27 to 2026-11-03 · 126 source reports.
| Range | Prices through |
|---|---|
| 1W | |
| 1M | |
| 3M | |
| 6M | |
| 1Y | |
| 2Y | |
| 3Y |
Research through 2026-11-03. Artifact validity does not establish factual accuracy or research freshness.