Reading

IBM-owned technology spend management firm Apptio announced on November 9, 2026, an expansion of its product portfolio with direct connections for leading frontier AI systems. These updates aim to help enterprises incorporate AI consumption data into broader technology financial management, addressing the challenge of fragmented AI costs and their disconnect from business value.

A June 2026 study by the IBM Institute for Business Value (IBV) found that 85% of surveyed technology executives reported lacking full visibility into real-time AI spending. With costs spread across model providers, cloud infrastructure, applications, and teams, organizations need a clearer picture of what AI initiatives cost and which are delivering business value.

Key components of the update include:

1. Native AI System Connections: Apptio introduces direct connections to major frontier AI systems, capturing model consumption and usage data (such as token counts) and merging it with traditional cloud and IT spend analysis.

2. AI-Assisted Analysis and MCP Servers: New Model Context Protocol (MCP) servers allow external AI tools to access financial, operational, and organizational data. Users can query cloud and AI spend via natural language to quickly identify drivers and risks.

3. AI Value & ROI Capabilities: The IBM Apptio platform now includes 'AI Value & ROI' features designed to connect technology investments with measurable business outcomes, supporting driver-based planning to improve forecasting of variable spend.

4. Enhanced Kubernetes Network Cost Visibility: IBM Kubecost gains detailed visibility into network costs generated by Kubernetes environments, including AWS NAT Gateway ingress/egress and cross-zone/region traffic, helping teams understand the drivers behind containerized workloads.

"AI investment is growing faster than many organizations can effectively govern and evaluate," said Eugene Khvostov, Chief Product Officer at IBM Apptio. "We're expanding these capabilities... to give teams the technology financial intelligence they need to understand the economics of AI at the most granular and real-time level."

Availability: AI system connectors are generally available, with additional sources planned for Q4. IBM Cloudability support for AI Commitments arrives in Q3, while AI Unit Economics are generally available with enhancements in Q4 2026 and incremental support through 2027.

US software company Adobe (Nasdaq: ADBE) and The Walt Disney Company announced a multi-year strategic alliance on November 9, 2026, at the Adobe MAX conference. The partnership aims to advance creativity, innovation, and storytelling through new technologies, marking a deepening integration of generative AI into the operations of two companies with a longstanding relationship.

According to Adobe's official press release, the core of the alliance involves integrating Adobe's AI technology into Disney's production pipelines. Adobe will serve as an "Official AI Innovation Partner" for Disney, providing early access to its latest technologies, including Adobe Firefly (generative AI for creatives), Adobe Firefly Foundry (proprietary AI models), Premiere, and Photoshop, along with dedicated engineering and research resources.

A key component is Adobe Firefly Foundry, which allows Disney to train customized proprietary models based on its own intellectual properties, such as character attributes, environments, and visual styles. These models are designed to generate commercially safe and production-ready outputs, thereby accelerating creative work and enabling more projects. Walt Disney Imagineering R&D is already utilizing this technology, for instance, using a sketch-to-image model to transform hand-drawn concepts into fully rendered 2D concept art.

The collaboration also extends to StudioLAB (Disney Studios' innovation hub) and ESPN Edge (sports storytelling technology center), exploring how AI can enhance fan engagement and content delivery. While specific financial terms or valuation impacts were not disclosed, this move demonstrates how major media conglomerates are balancing efficiency gains with intellectual property protection through proprietary AI models, offering a significant reference point for the commercialization of AI in the content industry.

US investment bank Oppenheimer upgraded Advanced Micro Devices (NASDAQ: AMD) from "Perform" to "Outperform" on November 9, 2026, setting a new price target of $600. The firm cited strong data center performance and growing demand for AMD's artificial intelligence (AI) chips following the company's latest earnings call as key reasons for the more bullish stance.

After discussions with AMD CFO Jean Hu and Vice President of Investor Relations Matt Ramsay, Oppenheimer highlighted server CPUs and initial shipments of the MI455 and Helios products as primary drivers behind approximately doubled data center growth. The analysts expect data center revenue to accelerate further through 2027, with GPU sales eventually surpassing EPYC server CPU revenue. AMD management has projected overall revenue growth exceeding 80% in 2027.

Oppenheimer estimates that AMD will reach approximately 3 gigawatts of data center capacity in 2027 and 4 gigawatts in 2028. Consequently, the firm predicts data center revenue will increase by 135% next year and eventually account for about 85% of total sales. This outlook is supported by AMD's plan to maintain an annual product release schedule, with the MI500 expected in 2027 and the MI600 following in 2028 to keep the company competitive in the AI accelerator market.

On the demand side, Oppenheimer expects OpenAI and Meta (NASDAQ: META) to support data center growth through the first quarter of next year, followed by another batch of orders from Anthropic in later quarters. Additionally, AMD shares could benefit from increasing demand for high-performance computing and related infrastructure investments. The five largest cloud service providers are expected to spend approximately $1.3 trillion on capital expenditures in 2027, up from $800 billion in 2026.

Other Wall Street firms have also responded positively to AMD's Q3 results. DA Davidson raised its price target to $600 while maintaining a "Buy" rating, StoneX reiterated "Buy" with a $685 target, TD Cowen maintained "Buy" with a $675 target, and Baird reiterated "Outperform" with a street-high $1,250 target. These moves reinforce the growing importance of AMD's business model, particularly its data center efforts, and the market's optimism about its success.

Does the relocation of gold reserves from the United States signal a weakening of the U.S. dollar's reserve status? In a FEDS note published in November 2026, Federal Reserve staff member Colin Weiss provides a data-driven negative answer by reviewing the repatriation cases of three euro area central banks—Germany, the Netherlands, and France—since 2013. The analysis shows that while these central banks transferred gold out of U.S. vaults, the share of the U.S. dollar in their foreign exchange reserves did not experience a substantial decline.

This finding is significant because it distinguishes between two different behavioral patterns. For Russia and Turkey, gold repatriation occurred simultaneously with a reduction in U.S. dollar holdings: Russia accumulated gold and reduced its dollar positions in the 2010s due to concerns over U.S. financial sanctions; Turkey moved gold stored in the United States and the United Kingdom back to its domestic territory in 2018 amid geopolitical tensions, while significantly lowering its share of dollar reserves. For these two countries, the core motivation for repatriating gold was to evade sanctions and asset freeze risks, and since U.S. dollar assets themselves are tools of sanction, the two decisions were logically consistent.

The situation in the euro area is different. Between 2013 and 2017, Germany transferred approximately one-fifth of its gold held in the United States to domestic vaults, maintaining its share of dollar foreign exchange reserves at around 90% during this period, and currently exceeding 80%. In 2014, the Netherlands repatriated about two-fifths of its U.S.-custodied gold to Amsterdam, with its dollar share remaining stable at approximately 83%. From 2025 to early 2026, France moved all its U.S.-held gold back home, with its dollar share stabilizing at 60%. Weiss notes in a footnote that the official statements from these central banks regarding gold relocation do not list sanctions or expropriation risks as primary motivations.

Weiss explains why the custodial location of gold and the share of dollar reserves are two decisions that can be decoupled. As a reserve asset, gold can help evade financial sanctions and asset seizures, but this advantage only holds if the gold is not stored within the jurisdiction of the government imposing sanctions. Meanwhile, the share of the U.S. dollar in foreign exchange reserves depends on sanctions risk, but also on the size and strength of the U.S. economy, as well as the unmatched depth and liquidity of U.S. financial markets. For euro area central banks, the economic benefits of allocating reserves to U.S. dollar assets appear to outweigh potential costs, meaning that even after relocating gold, there was no trigger for reducing dollar holdings.

It should be noted that this FEDS note represents the personal academic analysis of Federal Reserve Board staff and does not reflect the policy position of the Federal Reserve. Weiss himself acknowledges that the Netherlands did see a 10-percentage-point decline in its dollar share when it moved gold to London in 2026, although this level remains higher than the 2018 low and the dollar still dominates Dutch foreign exchange reserves. Additionally, a footnote mentions inconsistencies between the Netherlands Bank's annual report and the IMF's IRFCL monthly data after 2022, advising readers to exercise caution when interpreting specific figures.

According to a report by technology news outlet ZDNET, South Korean electronics giant Samsung Electronics is planning to implement a tiered subscription model for its built-in "Galaxy AI" smartphone features. This move suggests that power users who rely heavily on AI tools may soon have to pay for functionalities that were previously offered for free.

The proposed plan, dubbed "Galaxy AI Premium," is expected to start at $4.99 per month or $47.99 annually. Subscribers would receive usage limits four times higher than the free tier and gain access to exclusive high-compute features such as AI-generated videos, Bixby Action, and Daily Briefing. Basic tools like Photo Assist, Now Brief, and Creative Studio will remain free but will no longer offer unlimited usage; once the allotted free credits are exhausted, users will need to purchase top-up packages or subscribe to continue using these features.

The changes are reportedly planned for the upcoming Galaxy S27 series and will later expand to older devices, including the S24 series and newer foldable phones, via One UI updates. For occasional users, the impact may be minimal, but for those deeply integrated into daily AI workflows, this introduces a new recurring software cost. Samsung has not yet announced an official release date, though its flagship Unpacked events typically occur in January or February.

Google Cloud announced on November 10, 2026, that "Dynamic Slicing" is now generally available on its Kubernetes Engine (GKE). The feature addresses efficiency bottlenecks caused by the tight coupling between physical hardware configuration and logical compute topologies when running large-scale AI workloads on Tensor Processing Units (TPUs).

In traditional setups, managing TPU capacity required creating separate, static nodepools for each slice topology. This meant platform teams had to manually delete and recreate nodepools as workload needs changed, increasing complexity and creating large failure domains where a single hardware fault could crash an entire slice. GKE Dynamic Slicing decouples physical hardware provisioning from logical slice formation, leveraging TPU Optical Circuit Switching (OCS) technology for flexible topology management.

The feature operates in two modes: "super-slicing" and "sub-slicing." Super-slicing dynamically connects multiple TPU cubes (each consisting of 16 VMs) at scheduling time to form larger slices exceeding a single cube's size, catering to massive training demands. Sub-slicing partitions a single pre-provisioned TPU cube into multiple independent smaller slices, allowing concurrent, isolated execution of different workloads within the same physical cube. This eliminates the traditional "delete-and-create" nodepool cycle, enabling zero-downtime topology agility.

According to Google Cloud, compared to static nodepools, dynamic slicing reduces AI workload startup times by up to 5x, as the system instantly assembles logical slices from pre-provisioned nodepools instead of waiting for new deployments. Additionally, because failures are isolated to individual TPU cubes and unhealthy cubes can be swapped in one to two minutes, workload recovery speed improves by 4.5x. The feature also supports provisioning partially healthy nodepools, further enhancing overall fleet resource utilization.

For enterprises relying on TPUs for large model training and inference, this update signifies that the infrastructure layer can better adapt to rapid changes in model architectures, reducing operational complexity and improving fault tolerance. However, the cited performance improvements are based on Google Cloud's official testing, and actual results may vary depending on specific workload characteristics and environmental configurations.

Ride-hailing giant Uber announced on Monday the official launch of "Uber Parking," a new service allowing users to reserve and pay for parking in major U.S. cities. This move comes approximately nine months after the company acquired the parking app SpotHero. According to TechCrunch, the acquisition gave Uber access to a network of 13,000 garages, lots, and valet services across more than 400 cities.

While the service initially targets commuters driving their own vehicles, Uber explicitly stated its long-term strategy involves leveraging these parking resources to support dozens of autonomous vehicle companies it has partnered with. The company noted that these spots could serve as "staging and waiting locations" for autonomous vehicles, aiming to reduce congestion and "deadheading"—the miles a vehicle travels without a paying passenger or cargo. For the autonomous vehicle industry, reducing empty miles is a critical bottleneck for improving unit economics.

Uber CEO Dara Khosrowshahi indicated earlier this month during the third-quarter earnings call that commute is one of Uber's largest categories. He stated, "To be able to offer commuters an Uber to work, a pass to avoid surge, and parking if they want it when they want to drive is a great example of us building out our inventory of services to the commuter who is a top customer for us." Uber Parking is currently live in 15 U.S. cities, including Atlanta, Austin, Boston, and Chicago.

The service is deeply integrated into Uber's existing ecosystem. Users can access it directly from the app's home screen, entering their destination and time to filter options like self-parking, valet, or garages on an interactive map, with all pricing shown upfront. Uber One members (over 55 million as of September 30) can earn 10% back in credits on eligible parking reservations, which can later be used for rides and food delivery orders. This strategy continues Uber's path of increasing user stickiness and platform usage frequency by expanding product offerings such as hotel bookings, car rentals, and public transit connections.

AI voice technology company ElevenLabs launched ElevenAgents Campaign Manager on November 9, 2026. This new product is designed to help teams achieve outcomes in areas such as sales, retention, patient care, collections, customer service, and recruiting. Users provide a list of contacts and a desired outcome, such as a qualified lead or confirmed appointment, and the system intelligently runs the campaign, following up with each contact over time and carrying context between touches.

Unlike traditional dialers that rely solely on lists, Campaign Manager automatically schedules retries based on call results, such as no answer or voicemail, or redirects calls to other individuals on an account if necessary. For example, in a telecom renewal scenario, if the first call goes unanswered, the system schedules a retry within the set calling window. If the second call reveals that a daughter manages the family plan, the agent confirms her identity and proceeds to call her to complete the renewal. The product has been in beta with customers in insurance, banking, and e-commerce, handling daily lead lists, credit card qualification, and abandoned cart follow-ups.

To maintain control over compliance, Campaign Manager includes built-in features such as per-campaign calling windows, attempt limits, do-not-call lists that update instantly when a contact asks to stop, and mandatory disclosures in opening prompts. Every attempt, callback, and redirect is logged for team review. The system also integrates with existing ElevenAgents guardrails, checking responses in real time, detecting AI agents on the other end of calls, and redacting sensitive details from transcripts.

The product is now available directly within the ElevenAgents platform, utilizing existing agents and phone numbers so no infrastructure rebuild is required. Contacts can be enrolled via CSV upload or API. It currently supports phone calls, WhatsApp voice, and messaging, with email and SMS coming soon. Jeremy O’Sullivan, CTO of The AA Ireland, stated that using the tool increased their daily lead processing rate from 50-60% to 100%, with better voicemail detection than their previous system.

Venture capital firm Lightspeed Venture Partners announced on November 9, 2026, that it has led the Series C funding round for Nominal, a software company specializing in hardware test data. Alongside the financing, Nominal launched three new AI agents—Analyst, Execute, and Plan—intended to bring artificial intelligence into traditional hardware engineering workflows.

Hardware engineering has long suffered from "data silos." Unlike software development, where code changes can be tested instantly, data from rocket or satellite tests often resides on offline test stands, USB drives, or individual engineers' Python scripts, making it inaccessible to modern AI models. Nominal's core strategy involves building a unified data layer that aligns high-rate sensor data, video, and logs on a common timeline, creating a foundation upon which AI agents can operate.

The newly released agents target different stages of the testing process: Analyst visualizes data, reviews specifications, and debugs failed tests; Execute controls instruments to run tests and automatically aborts them if anomalies are detected; and Plan helps engineers decide which parameters to test next to maximize information gain. Lightspeed noted that approximately 600,000 engineers in the US work in industries driving most R&D, representing about $100 billion in annual labor costs. Even optimizing a fraction of this time presents significant economic potential.

Nominal's customer roster includes aerospace and defense technology companies such as Anduril, Shield AI, and Hermeus. In its announcement, Lightspeed claimed that Nominal helped Hermeus achieve hypersonic flight faster than any program since the 1960s and secured a sole-source contract worth up to $53 million from the Air Force Test Center. However, these specific outcomes rely primarily on vendor and client statements, with no independent third-party efficacy assessments currently available.

US technology company Meta announced on November 9 that it is expanding its AI-powered age-estimation technology on Instagram to a global scale. The move aims to more accurately identify users suspected of being teenagers and place them under the strict protections of "Teen Accounts," even if those users provided an adult date of birth during registration.

According to TechCrunch, Meta's age-estimation tools use AI to analyze contextual clues across entire profiles, such as birthday celebration posts, mentions of school grades, and content across various formats including posts, comments, bios, and captions. Additionally, the company examines visual cues from photos and videos, including a person's height or bone structure. Once the system suspects a user is a teen, the account is automatically classified as a "Teen Account," triggering additional safeguards such as restricting direct messages to people already followed or connected, hiding harmful comments, and setting accounts to private by default.

This global rollout comes amid significant legal pressure on Meta. In August, Meta reached settlements worth up to $18 billion with nearly every US state and territory to resolve lawsuits alleging its platforms harm teens. As part of these agreements, Meta committed to improving how it identifies teens and children under 13. "That’s why as part of our agreement with 52 bipartisan attorneys general... we’re investing in even stronger technology to proactively catch accounts that may belong to children under 13," Meta wrote in a blog post.

Beyond Instagram, Meta stated it is continuing to roll out the technology on Facebook, with a similar global rollout planned for Threads later this year. For parents in the US, Instagram will display an in-feed video explaining Meta's approach to ensuring their teens are in age-appropriate protections. This shift marks a transition for social media platforms from relying on self-reported ages to employing complex AI behavioral analysis and biometric inference to meet regulatory demands.

According to a report by the Financial Times, Apple Inc. has shifted its primary tax haven status from Ireland to Singapore, as revealed in its latest annual filing with the U.S. Securities and Exchange Commission (SEC). This marks the first time Apple has provided such detailed country-level tax disclosures under new US accounting regulations.

The Financial Times estimates that Apple saved approximately $2.7 billion in taxes during fiscal year 2026 by booking profits in Singapore. Although Singapore's implementation of a new global minimum supplementary tax increased Apple's local tax burden by about $2.5 billion, the net savings compared to other jurisdictions remained substantial.

The filing shows Apple achieved a pre-tax income of $159.2 billion and paid $27.1 billion in income taxes for the fiscal year, resulting in an effective tax rate of 17.0%. Previously, Apple relied heavily on Ireland's low corporate tax rates, but international tax reforms have necessitated a restructuring of its global profit allocation.

For investors, this transparency offers a clearer view of how geographic tax strategies impact net income. While the FT estimates that up to $50 billion in profits flowed through Singapore entities, Apple has not independently confirmed this specific figure, only disclosing the tax effects required by the new standards.