Israel stocks lower at close of trade; TA 35 down 0.49%

Investing.com – Saudi Arabia stocks were lower after the close on Sunday, as losses in the Transport, Building & Construction and Media & Publishing sectors led shares lower.

At the close in Saudi Arabia, the Tadawul All Share declined 1.19%.

The best performers of the session on the Tadawul All Share were flynas Company SJSC (TADAWUL:4264), which rose 2.55% or 1.80 points to trade at 72.30 at the close. Meanwhile, National Industrialization Co (TADAWUL:2060) added 2.13% or 0.21 points to end at 10.09 and Wataniya Insurance Company (TADAWUL:8300) was up 1.92% or 0.25 points to 13.30 in late trade.

The worst performers of the session were Sustained Infrastructure Holding Company SJSC (TADAWUL:2190), which fell 5.75% or 1.88 points to trade at 30.82 at the close. CHUBB Arabia Cooperative Insurance (TADAWUL:8240) declined 5.72% or 1.36 points to end at 22.40 and Arabian Shield Coop Insurance Co (TADAWUL:8070) was down 5.25% or 0.62 points to 11.18.

Falling stocks outnumbered advancing ones on the Saudi Arabia Stock Exchange by 279 to 51 and 14 ended unchanged.

Shares in CHUBB Arabia Cooperative Insurance (TADAWUL:8240) fell to 52-week lows; down 5.72% or 1.36 to 22.40. Shares in Arabian Shield Coop Insurance Co (TADAWUL:8070) fell to 5-year lows; down 5.25% or 0.62 to 11.18.

Crude oil for February delivery was down 0.28% or 0.16 to $57.24 a barrel. Elsewhere in commodities trading, Brent oil for delivery in February fell 0.26% or 0.16 to hit $61.12 a barrel, while the February Gold Futures contract rose 0.35% or 15.30 to trade at $4,328.30 a troy ounce.

EUR/SAR was unchanged 0.01% to 4.41, while USD/SAR unchanged 0.01% to 3.75.

The US Dollar Index Futures was up 0.03% at 98.02.

Investing.com - Tech giants are scrambling to build out AI infrastructure, and given the amount of scrutiny on the recently announced large-scale datacenter commitments, analysts at Bernstein discuss what’s in a datacenter contract.

Contracts are generally bespoke, last 10-15 years with renewal options, and are typically difficult to cancel, the analysts at Bernstein said, in a note dated Dec. 10, as long term leases without easy way of backing out are essential for developers to finance. 

The hyperscalers are often the harshest negotiators who bring their own lease forms, specifications and customizations, but their leases also provide credit enhancement for developers to more easily obtain finance. 

Startup customers, on the other hand, need further credit enhancement tools such as letter of credit, cash deposit or guarantees from partners like Nvidia (NASDAQ:NVDA).

For datacenter developers, energy is the bottleneck, rather than location or supply chain issues. Hence, many are exploring alternative energy sources for longer-term build outs, Bernstein added, even if Microsoft (NASDAQ:MSFT) recently said that finding space rather than electricity was their biggest bottleneck. 

Hyperscalers, neoclouds and AI clouds might be committing 3-4x the capacity they need in the next 24 months to ensure reliable and timely supply; because realistically only a third of the contracts can be delivered on time, with another third delayed and the rest one third never built.

Large deals often have multiple tranches. Capacity is delivered in stages, often tied to power availability rather than construction speed.

Cloud players sign deals to mostly match their near-term capacity needs for the next 24 months, and they diversify multiple multi-tranche deals to manage and smooth out the timing of actual capacity received.

Multi-tranche deals seem to have more flexibility/uncertainty in timing and outcome in outer tranches, which naturally offset some of the “over-commitment”. All bespoke contracts can contain extension rights for delay in delivery acceptable, and can build in windows for client to put off start date for a few months. 

Although rare, there exist contracts with early termination options (for a fee), or termination based on specific delays as negotiated in the contract. Combined with the natural delays or failures from some projects, in aggregate from the cloud player’s perspective the amount and timing of capacity supply longer term could sound quite different from what’s initially signed.

Contract duration management could be a risk on both sides. For the cloud players especially neoclouds, the lease terms are at least 10-15 years yet their cloud deals are often 2-3 year contracts.

For the developers, buildings typically last longer than 15 years and they have traditionally seen high renewal rates, but that could be subject to change in the future.

Investing.com -- Morgan Stanley says investors may be underestimating the risk that AI could disrupt eCommerce and food-delivery platforms, as large language models begin to handle product discovery and ordering directly.

Instacart became the first grocery company to enable shopping inside ChatGPT. Morgan Stanley says such moves could accelerate questions about whether consumers might bypass retailer apps altogether. The feature will roll out in the coming weeks.

While the debate around “agentic AI” has been well-developed in classifieds, Morgan Stanley says that similar risks in eCommerce like Allegro, HelloFresh, Ocado, Zalando and food delivery such as Delivery Hero, talabat, Prosus, Jahez, have received far less attention.

Yet the core threat is the same: AI could redirect traffic, increase price transparency and weaken platform loyalty.

The analysts point out that AI concerns are already influencing stock calls. MONY, for example, was recently downgraded partly because AI could erode comparison-site economics.

HelloFresh, downgraded on rising competition, may also see substitution risks as AI reshapes how users plan meals and shop for ingredients.

For food-delivery firms, AI agents could end up selecting restaurants, comparing fees and placing orders, potentially compressing take-rates and shifting power to AI platforms. Several large operators “have yet to face the same degree of debate,” the analysts says.

Morgan Stanley expects AI-related disclosures to rise through 2026 as companies rush to show how they will avoid being disintermediated.

“This type of news flow will accelerate debate around what companies are doing to prepare,” the analysts write.

Investing.com -- Autonomous vehicles could reshape the U.S. transportation economy, with Bank of America arguing that driverless ride-hailing represents a “massive $1tr+ US market opportunity.” 

In a note this week, analyst Justin Post wrote that investors are increasingly focused on how AV technology will alter competition, pricing and profitability for mobility leaders such as Uber and Lyft.

According to BofA, U.S. drivers log 3 trillion miles annually, yet ride-hailing today captures only 1% of those miles. 

If AVs can lower per-mile ride costs closer to car-ownership levels, the bank believes adoption could expand dramatically. 

Post said that “AV ride-hailing could expand to 20% of annual US vehicle miles traveled (VMT) over the next 15 years,” implying a market worth $0.9–$1.2 trillion based on estimated per-mile pricing of $1.5–2.0.

For Uber, which BofA estimates has 70–80% U.S. market share, the long-term growth runway remains substantial even if new rivals such as Waymo, Tesla or Zoox enter aggressively. 

The firm sees “potential for Uber US mobility bookings to grow at a long-term CAGR of 13% to 20%” depending on share outcomes between 30–70%.

A key concern for investors is said to be whether cheaper AV rides will erode margins. But BofA believes Uber could maintain roughly 10% mobility margins even with a drop in ride prices to $2.00 from roughly $2.75 today. 

Under more favourable cost conditions, such as AV hardware falling to $45,000 and operating costs below 20%, the break-even price could fall to $1.60, the analyst notes.

BofA concludes: “We still see an impressive opportunity,” citing rising AV demand, strong network effects and viable economics for both riders and platforms.

Investing.com --  After hosting a webinar with Gunjan Shah, a former Senior Cloud Engineer, AI and Machine Learning at Google, analysts at Bernstein provided their thoughts on the amount of memory AI data centers require.

In its key takeaways, the firm explained that AI data centers require dramatically different amounts of memory depending on whether they are training or running models. 

Analyst Mark Newman says training demands “substantially more memory than inference,” because it requires storing model weights, activations, gradients, optimizer states and “frequent checkpoints.” 

Bernstein, citing the expert commentary, noted that even a medium-sized model can consume “~1TB of combined memory” during training. Inference, by contrast, needs far less, with storage limited to temporary tensors and KV caches.

Newman states that hyperscalers were caught off guard by the surge in AI adoption, triggering a sharp rise in memory demand and pricing. 

The resulting imbalance is said to have pushed up the cost of key components such as HBM and DRAM. 

However, the firm notes that improvements in model architectures, new quantization techniques and next-generation chips should help “manage memory demand over the long term” and support sustainability.

The note highlights storage as another bottleneck. A shortage of HDDs has pushed many operators toward SSDs. 

Bernstein adds that SSDs are “five to ten times more expensive” than HDDs, but companies are willing to absorb the cost to continue advancing their models. 

SSDs are also said to offer performance and efficiency advantages, including “lower operational costs, reduced power consumption, and minimal cooling requirements.”

Bernstein also points to purpose-built TPUs, which deliver “lower TCO, higher performance per watt and superior scalability,” though GPUs remain favored for rapid prototyping due to their mature ecosystem. 

Looking ahead, the firm says High Bandwidth Flash could become a critical new tier, offering terabytes of fast, non-volatile memory and lower energy needs for future AI workloads.

Investing.com -- Here is your Pro Recap of the top takeaways from Wall Street analysts for the past week.

InvestingPro subscribers always get first dibs on market-moving AI analyst comments. Upgrade today!

Ulta Beauty

What happened? On Monday, TD Cowen upgraded Ulta Beauty (NASDAQ:ULTA) to Buy with a $725 price target.

*TLDR: TD Cowen upgrades ULTA Buy. Needs TikTok, influencers, Gen Z.

What’s the full story? TD Cowen upgrades ULTA to Buy with a $725 target, betting that new CEO Kecia Steelman transforms this beauty box into something resembling actual growth. The analyst sees Steelman’s "operational expertise" fixing stores, supply chains, and whatever "cross-functional go-to-market alignment" means in corporate-speak.

The thesis: ULTA remains the dominant specialty beauty retailer with fat margins, 29,000 SKUs, and a loyalty program that actually works—essentially an asset manager for lipstick and mascara.

But here’s where it gets interesting. TD Cowen argues ULTA desperately needs to become TikTok’s beauty BFF, courting influencers while transforming stores into "wellness destinations"—because apparently selling makeup isn’t enough anymore. The analyst wants more "editorial marketing presence" and "community building," which translates to: stop being boring and start being where Gen Z discovers products.

They’re betting ULTA leverages its scale to lock down hot brands before competitors do, without letting them disappear into the retail abyss.

Classic retail turnaround story, except this time with better margins and shinier packaging.

Roblox

What happened? On Tuesday, Freedom Capital Markets initiated Roblox (NYSE:RBLX) at Hold with a $99 price target.

*TLDR: Roblox streams toward $9.5 billion bookings. Regulatory hellfire threatens growth momentum.

What’s the full story? Freedom Capital projects a 7.5x EV/bookings multiple on their FY27 $9.5 billion estimate, betting Roblox keeps printing money like a digital casino for kids. The analyst sees this gaming platform as the industry’s lone bright spot—democratizing video games while hoovering up 150+ million daily users and generating cash flows that make traditional publishers weep.

Q4 looks juiced thanks to gift card season, with bookings and DAUs rocketing 50% and 80% in FY25.

But here’s where the fairy tale gets ugly: multiple U.S. states are suing over safety failures, facial age verification is spooking users, and Freedom Capital spots "a bit soft" trends as the company cracks down on fake accounts. Margins face a squeeze as Roblox throws more money at developers and beefs up safety measures—essentially paying protection money against regulatory hellfire.

The analyst models FY25 bookings at $6.7 billion with 24.3% EBITDA margins, slightly above management’s suspiciously conservative guidance. Freedom Capital projects steady expansion through FY27, hitting $9.5 billion in bookings, though consensus dreams even bigger at $9.565 billion.

The real question lurking: whether Roblox can maintain its growth trajectory while navigating the minefield of child safety regulations and slowing user momentum.

Waters Corp.

What happened? On Wednesday, Wolfe Research upgraded Waters Corp. (NYSE:WAT) to Outperform with a $480 price target.

*TLDR: Wolfe upgrades undervalued WAT despite BD dilution. Management execution justifies higher valuation than today.

What’s the full story? Wolfe Research is upgrading WAT because apparently buying quality assets at a discount still works in this market as reports of value investing’s death have been greatly exaggerated (via Berkshire’s cash position overall being at record levels.).

The stock is trading at roughly 17x 2026 EV/EBITDA and 22.6x FCF, a 10-12% haircut to peers who frankly aren’t running half as tight a ship. Sure, the BD Life Sciences bolt-on might water down the organic growth story like cheap whiskey in a frontier saloon, but Wolfe’s analysts figure management knows what they’re doing.

The premium’s compressed from 25% to 15% post-deal, which seems about right for a company that just ate something that might give it indigestion. But here’s the thing: above-average core growth, beats on pro forma targets, gorgeous free cash flow, and a management team that actually delivers?

That cocktail’s worth more than today’s price, even if the BD acquisition makes the story marginally less pure.

Roku

What happened? On Thursday, Jefferies upgraded Roku (NASDAQ:ROKU) to Buy with a $135 price target.

*TLDR: Jefferies sees 20%+ growth, cleanest revision story.

What’s the full story? Jefferies smells blood in the water on Roku.

The analysts see the Street’s limp 15% year-over-year growth guess for 2026 Platform revenue as low-hanging fruit—try 20% or better once the DSP rockets fire (Amazon’s ramp, Trade Desk’s relentless march), political ad dollars flood in, subscriptions keep compounding, and that home-screen overhaul finally drops.

Management’s still swinging the cost-discipline axe hard enough that mid-single-digit opex growth can bankroll double-digit revenue for years. In a sector full of guidance cuts and excuse-making, Roku shapes up as the cleanest upward revision story on the board heading into 2026.

Citigroup

What happened? On Friday, JPMorgan upgraded Citigroup (NYSE:C) to Overweight with a $124 price target.

*TLDR: JPMorgan upgrades Citi, believes transformation story. Expects everything to be fixed eventually, somehow.

What’s the full story? JPMorgan throws Citigroup a lifeline, upgrading the perpetual turnaround story to Overweight because—get this—they believe in transformation. The analysts figure Citi’s revenue concentration makes it the perfect casino bet on "solid economy and strong markets," which surely nothing could disrupt.

Here’s the punchline: JPMorgan expects Citi to simultaneously fix its efficiency ratio, satisfy its regulatory overlords on those pesky consent orders, eliminate stranded costs like some corporate Marie Kondo, and somehow monetize that albatross DTA. All while outperforming peers on RoTCE—because if there’s one thing markets reward, it’s a bank promising to get its house in order tomorrow.

The analysts seem convinced these improvements will "continue over time," that delightfully vague phrase meaning somewhere between next quarter and the heat death of the universe.