Intel Soars 9% Then Plunges 6% in One Day, 300 Billion Dollar Boost Fails to Hold Off Profit-Taking

Publicado a 2026-09-11Actualizado a 2026-09-11

Resumen

Intel fell about 6% after a previous-day surge of 9%, as the market reassessed whether factors such as the share offering, Nvidia's stake, and CPU price increases are sufficient to support its valuation. Piper Sandler assigned a Neutral rating with a $110 price target.

Profit-Taking Weighs on Intel

Intel's stock fell approximately 6% in early trading Thursday, following a 9% surge in the previous session. With the price retreating to around $100, investors began reassessing whether the multiple catalysts driving the recent rally were sufficient to support the current valuation.

The previous rally involved the issuance of over $20 billion in common stock, Nvidia's disclosure of roughly $30 billion in holdings, and reports that Intel might raise CPU prices for a third time in 2026. After these developments converged, short-term capital chose to take profits.

Analysts Remain Cautious on Valuation

Piper Sandler initiated coverage on Intel on September 9 with a Neutral rating and a $110 price target. The firm believes that while there is genuine server CPU demand driven by agentic AI, supply constraints could persist until 2029 or 2030.

The sell-side market is still dominated by cautious views, with 32 Hold ratings and 13 Buy ratings. The consensus analyst price target offers only limited upside compared to recent prices, serving as an anchor for short-term sentiment.

Price Hikes and Financing Need to Materialize

CPU price increases could improve margins but will also test customer demand and the competitive landscape. Meanwhile, large-scale equity financing helps support investment plans but brings dilution pressure; the market needs to see capital translate into capacity and revenue.

Following its significant surge, Intel's single-day pullback does not necessarily mean the end of its transformation narrative, but the volatility reflects that good news has already been priced into the valuation. The focus going forward will be on the execution of price increases, improvements in server supply, and capital efficiency.

Lecturas Relacionadas

DATA Network joins Sui and Walrus to scale the supply chain for AI training data

The DATA Network, which operates the Trace onchain registry for rights-cleared AI training data with over 250 million verifiable receipts, is joining the Sui blockchain ecosystem and adopting Walrus as its default storage layer. This integration addresses a critical need for the next generation of AI models, which require real-world, consented data with clear provenance due to regulatory requirements and legal challenges. The open internet has been largely exhausted as a data source. Trace registers the origin and usage rights of data from contributors—such as personal dashcam footage or medical notes—and handles payments to them, processing over 3 million receipts daily. Sui's object-centric model allows datasets and their licenses to be managed onchain, with fast, parallel transaction processing. Walrus provides efficient, secure storage for large files, ensuring data is only accessible to licensed buyers. By integrating with Walrus, Trace will be able to store an immutable, encrypted copy of the original data file itself, strengthening provenance guarantees. As part of the migration, the DATA Network will be upgraded to run natively as a protocol on Sui, and its independent chain will be retired. The $DATA token will transition to the Sui protocol on a one-to-one basis. Sui Foundation is making a $30 million strategic investment in the DATA Network, becoming its largest institutional stakeholder. Leaders from Mysten Labs (original contributor to Sui), PIP Labs, and DATA Foundation highlighted that the combination provides a comprehensive solution for storing, licensing, and paying for the high-quality, consented data essential for advancing AI, with the scale and integrity proofs the industry demands.

TheNewsCryptoHace 10 min(s)

DATA Network joins Sui and Walrus to scale the supply chain for AI training data

TheNewsCryptoHace 10 min(s)

ICP Cloud Engine Heats Up: "20% of Revenue Used for Burn", How Far is it From Token Revaluation?

The ICP community is discussing Cloud Engines, a product enabling enterprises to run private subnets on the Internet Computer. It is linked to the "Mission 70" deflationary plan, where node providers allocate 20% of Cloud Engines revenue to buy and burn ICP tokens, aiming to reduce ICP's inflation rate from 9.72% to 2.92%. This narrative connects enterprise adoption to token scarcity and potential price appreciation for ICP, which is down 99.6% from its peak. Cloud Engines functions as a "custom data center," offering businesses dedicated, sovereign infrastructure without vendor lock-in, using the same underlying protocol as the public ICP network. Key selling points include data sovereignty and hardware choice. However, critical features like SEV-encrypted nodes (for enhanced security) and operator-controlled upgrades are still on the roadmap, not yet live. Early adopters include Pakistan's national cloud engine and a UNDP pilot. The "Mission 70" goal relies on two pillars: supply-side reductions (like cutting node rewards) and demand-side increases in token burns from Cloud Engines revenue. While supply-side changes can be enacted via governance, the demand-side depends entirely on market adoption. Current daily burn rates are far below the levels needed to achieve the 26% deflation target attributed to increased burns. Analysts remain cautious, viewing the 20% burn mechanism as a forward-looking narrative. The product is in early access, and significant, scalable enterprise adoption is unproven. Near-term price catalysts are more likely to come from implementable supply-side governance proposals. The Cloud Engines story represents a longer-term bet on ICP capturing a segment of the enterprise cloud market, with token revaluation contingent on the product generating substantial, quantifiable burn data.

marsbitHace 30 min(s)

ICP Cloud Engine Heats Up: "20% of Revenue Used for Burn", How Far is it From Token Revaluation?

marsbitHace 30 min(s)

Bank of America's Hartnett: Interest Rate Hikes Become the Biggest Positive, Bond-Ending Bubble Logic Is Spreading

In a recent note, Bank of America strategist Michael Hartnett argues that aggressive central bank tightening, including a potential Fed hike in October, could paradoxically be the most bullish near-term catalyst for risk assets. His core thesis is "the bond bubble is bursting." While stocks remain resilient due to strong US data, solid earnings expectations, and robust AI-related capital expenditure plans, bonds are already showing severe stress, with government bonds down 4.4% year-to-date. The logic is that a proactive Fed hike could help anchor runaway long-end yields (the 10-year recently touched 5.36%), preventing a disorderly surge toward 6%. Without such action, the pressure from rising term premiums could continue unabated. The damage from higher rates is already visible in the most rate-sensitive areas: South Korean retail brokers, French government bonds (with the France-Germany spread hitting 140 bps, reminiscent of the 2012 crisis), and CCC-rated credit. Hartnett highlights that the $8 trillion in money market funds is unlikely to flow into equities without a clear shift to monetary easing. Furthermore, the upcoming US midterm elections pose a key risk; a Democratic sweep (currently a 64% probability) could trigger a >10% equity sell-off due to fears of higher taxes and regulation. Near-term strategy involves positioning for a peak in yields (e.g., long 30-year Treasuries) and risk contraction ahead of the elections, while maintaining exposure to commodities and EM assets. Key levels to watch include 10-year yields at 6%, the France-Germany spread at 150 bps, and a Bull & Bear Indicator below 8 for a potential risk rebound.

marsbitHace 34 min(s)

Bank of America's Hartnett: Interest Rate Hikes Become the Biggest Positive, Bond-Ending Bubble Logic Is Spreading

marsbitHace 34 min(s)

ETF Watch: ETH Outflows for Nine Consecutive Days, SOL Ends Fourteen-Week Inflow Streak, Combined Weekly Outflows from Three Major Spot ETFs Total Approximately $1.25 Billion

U.S. spot cryptocurrency ETFs saw significant outflows last week. According to The Block citing SoSoValue data, spot Ethereum (ETH) ETFs experienced net outflows of $542.1 million, marking their worst week since January and continuing a nine-day streak of withdrawals. Spot Solana (SOL) ETFs ended a record 14-week inflow streak with a weekly net outflow of $24.8 million. Combined with Bitcoin (BTC) ETF redemptions, the three major categories saw approximately $1.25 billion in total outflows for the week. BlackRock's iShares Ethereum Trust (ETHA) led the Ethereum outflows, accounting for $477 million or roughly 88% of the category's total. The largest Solana ETF, Bitwise's BSOL, saw its largest weekly outflow on record at $20.9 million. Bitcoin ETFs saw net outflows of $681.1 million, ending a three-week inflow streak. This shift indicates a broad cooling of investor sentiment through the ETF channel, moving from a pattern of "BTC inflows and SOL gains" to simultaneous withdrawals across major digital asset funds.

marsbitHace 34 min(s)

ETF Watch: ETH Outflows for Nine Consecutive Days, SOL Ends Fourteen-Week Inflow Streak, Combined Weekly Outflows from Three Major Spot ETFs Total Approximately $1.25 Billion

marsbitHace 34 min(s)

Silicon Valley AI Status: Whoever Has the Compute is the Boss

The Silicon Valley AI Power Crunch: Compute is King Silicon Valley's race for AI dominance has devolved into a desperate scramble for computing power. The situation has become so extreme that even top executives are resorting to unusual measures. Anthropic's CEO Dario Amodei was reportedly turned down by Meta's Mark Zuckerberg when seeking additional GPU resources, leading to a deal where SpaceX acts as a broker, reselling Nvidia GPUs to Anthropic. This venture reportedly generated $2.561 billion in quarterly AI revenue for SpaceX. The shortage is reshaping the industry's dynamics, creating a new "compute economy" where GPU access is the ultimate currency. This scarcity is causing internal strife at major firms and altering competitive landscapes. A high-profile example is Noam Shazeer, a key AI scientist whom Google re-hired for $2.7 billion, only for him to leave for OpenAI reportedly due to insufficient compute resources for his team. Similar frustrations over resource allocation have been cited within Google and DeepMind. For AI startups, the challenge is existential. Cloud providers are now demanding multi-year contracts with large upfront payments, making flexible, on-demand access nearly impossible. This has spurred innovative workarounds, including a new market for "global compute prospecting"—companies scavenging for underutilized server capacity worldwide—and venture capital firms like Y Combinator and Radical Ventures directly acquiring and allocating GPU clusters for their portfolio companies. The root causes are complex: explosive growth in AI model complexity and the shift from simple chatbots to persistent "Agent" systems dramatically increase compute demands. While tech giants are building new data centers, supply cannot keep pace with immediate needs. Rental prices for critical chips like the Nvidia H100 have surged nearly 40% in months. The consequence is a consolidation of power. Compute access is becoming the definitive barrier to entry, significantly strengthening large companies with their own infrastructure and creating a lucrative "picks and shovels" market for those who control the hardware supply. In today's Silicon Valley, compute isn't just a resource—it's the foundation of survival and competitive advantage.

marsbitHace 1 hora(s)

Silicon Valley AI Status: Whoever Has the Compute is the Boss

marsbitHace 1 hora(s)

Trading

Spot