Author: Merrel Briona

After mounting pressure on Nvidia’s board to fire its scandal-plagued chief executive, the company’s CEO Jensen Huang is officially resigning.

According to a press release published early Thursday morning before the opening bell, Nvidia announced its “CEO transition” with Huang resigning as “President, Chief Executive Officer, and a member of the Board, effective immediately.”

In his stead, the AI chipmaker’s current executive VP and CFO Colette Kress will serve as interim CEO.

The New York Times revealed late last month that Huang had an extramarital affair with a suspected Chinese spy. This week, the newspaper revealed he had a second affair with Nvidia health executive Kimberly Powell.

Huang’s identity is practically inseparable from Nvidia, which he founded back in 1993 with the support of the wife he later cheated on.

As such, reactions online were explosive.

“It’s official: diva down,” wrote journalist Paris Martineau on social.

The company’s stock is likely to be in turmoil when markets open later this hour. The company’s shares are already down over 11 percent in pre-market trading.

More on Nvidia: We Are Now Learning in Real Time Whether the Stock Market Cares About an AI Executive Admitting to an Affair With an Alleged Chinese Spy

NEW YORK, Nov 12 : Warby Parker is entering the smart glasses market in partnership with Google and Samsung, betting that demand for the artificial intelligence wearables will balloon in the coming years.

But the New York-based fashion brand, known for its affordable eyewear, is also seeking to head off privacy concerns that have weighed on existing producers of the glasses, most of which embed small cameras allowing wearers to record their surroundings without taking out their phone.

The Warby Parker glasses are broadly similar to Meta smart glasses already available from Ray-Ban and Oakley maker EssilorLuxottica, which have faced regulatory scrutiny from Norway to Australia. Also like the Meta glasses, they will connect to an AI assistant, in this case Google’s Gemini, for everything from directions to language translation.  

The big difference from existing versions on the market is the privacy framework embedded in the glasses from the outset, Warby Parker Co-CEO Neil Blumenthal told Reuters.

Alphabet’s Google will not use visuals captured with the glasses to train its AI. The glasses will bar facial recognition of bystanders. And a small white notification light on the front of the glasses automatically turns on when a user starts recording video.

Shoppers will start placing pre-orders from Warby Parker on Thursday. Ray-Ban’s Meta glasses still dominate, representing 81.3 per cent of AI glasses shipments in the first half of this year, according to tech research group Omdia.

But brands from Warby Parker to luxury group Kering — poised to release glasses also in partnership with Google — are looking to gain a foothold. Global AI glasses shipments more than doubled in the first half of 2026 from a year ago, though growth has slowed since the second half of last year, Omdia data shows.

Meta’s AI glasses also include a camera notification light on the front, a Meta spokesperson said, adding that input from safety experts helped inform the privacy features on its glasses.

CATEGORY CONFRONTS PRIVACY RISKS

Optimism that Warby Parker’s “intelligent eyewear” plan will prove to be a major growth engine has helped its shares rally by nearly 40 per cent this year, outperforming a 10 per cent gain in the S&P 500.

The company expects an addition of several percentage points to fourth-quarter revenue growth from the new wearable technology, executives said on an earnings call earlier this month.

“As long as the tools provide utility to customers, then we’ll be successful,” Blumenthal said.

While the core optical backdrop remains soft, smart glasses are poised to be an “increasingly important driver” of Warby Parker’s growth outlook, analysts at TD Cowen wrote in a research note last week. 

Still, concerns about privacy could slow their pace of growth. One of Britain’s biggest pub chains has restricted the use of the frames, Norway’s government is planning a temporary ban on the use of AI glasses in certain locations, and Australia is considering barring the use of camera-equipped smart glasses in government workplaces.

GOOGLE AIMS TO REBUILD TRUST

Juston Payne, who leads extended reality products at Google, acknowledged that public skepticism could be a risk to growth.

“This is an area where, understandably, trust was lost,” he told Reuters at a demo for the Warby Parker glasses in New York.

Google says it will embed extensive privacy guardrails into the devices to prevent tampering with the recording indicator light; redact people’s faces from video footage before it is sent to Gemini; and ban targeted advertising on the glasses.

“We are going to change this conversation,” Payne said. “We are trying to raise the bar.”

But it remains to be seen whether those measures will ease concern about widespread surveillance, if the glasses become ubiquitous.

“I’m certainly happy to hear that (video footage is) not automatically fed into AI systems for training,” said Calli Schroeder, who leads the AI and human rights program at the Electronic Privacy Information Center, an advocacy group. “But that still leaves open: a stranger has footage of my kid on their phone. That’s concerning no matter how you phrase it.”

Analyst(s): Keith Kirkpatrick
Publication Date: November 12, 2026

What Is Covered in This Article:

• Adobe Creative Agent and the new AI assistants across Firefly and Creative Cloud
• Native MCP support (in preview) and connectors that let third-party AI agents operate Adobe desktop apps
• Firefly’s expansion into a multi-model AI creative studio, including Firefly Design, Builder, and Graph
• GenStudio updates for enterprise content supply chains, including CX Enterprise Coworker and Brand Intelligence
• The CEO handoff from Shantanu Narayen to Anil Chakravarthy

The Event—Major Themes & Vendor Moves: Adobe MAX 2026 opened November 10 in Miami Beach and runs through November 12, with more than 300 speakers and keynotes streamed online. The opening keynote doubled as a handoff: Shantanu Narayen, closing out 29 years at Adobe, cited more than a billion monthly active users across Adobe products before passing the stage to incoming CEO Anil Chakravarthy.

Chakravarthy’s message was focused on providing customers with choice. Creators who want nothing to do with AI get non-AI tooling such as dynamic text in Illustrator and a rebuilt Color Mode in Premiere. Those who want AI get the Adobe Creative Agent, which now powers AI assistants in Firefly, Premiere, After Effects, Illustrator, InDesign, and Frame.io, plus third-party models from Google, Kling, Runway, Luma, Flux, OpenAI, ElevenLabs, and the recently acquired Topaz Labs, all under one subscription.

The most consequential announcement was native Model Context Protocol (MCP) support, now in preview, across Adobe’s desktop creative tools, demonstrated with OpenAI and Anthropic models operating Photoshop and Illustrator on stage at Adobe MAX 2026. Adobe connectors already run inside Anthropic, Google, OpenAI, and Slack surfaces, with Microsoft Copilot coming soon.

Firefly expanded into what Adobe calls an AI creative studio, adding Firefly Design (beta), the no-code Firefly Builder (private beta), and Firefly Graph (public beta) for reusable node-based workflows. For enterprises, Adobe is bringing CX Enterprise Coworker into GenStudio, adding Brand Intelligence to enforce brand standards, and announced a multi-year alliance with The Walt Disney Company. Adobe also said nearly 70% of professional Photoshop users regularly use its AI features.

Adobe MAX 2026: Adobe Ready to Let Other AI Agents Run Its Apps

Analyst Take—Adobe MAX 2026 Shows Adobe Ceding the Front Door: Adobe MAX 2026 delivered plenty of features, but the strategic signal is that Adobe is willing to let someone else own the interface. Native MCP support (in preview) and connectors into Claude, ChatGPT, Gemini, and soon Copilot mean a user can direct Photoshop or Illustrator without opening either application. Adobe is positioning its apps as the trusted execution layer behind whatever agent a customer prefers.

The counterweight to this approach is Adobe’s professional base. Serious creatives are unlikely to abandon Photoshop, Premiere, or Illustrator just because an agent can drive them; they will keep working inside the core apps for the pixel-level precision and control their clients pay for. Adobe’s own numbers point the same way: Photoshop shipped 77 new features across 32 releases this year, and Adobe said its most advanced users are the heaviest adopters of its AI features.

That loyalty is Adobe’s real defensive moat, and it reaches beyond the creative team, because creative professionals often shape which tools the enterprise licenses and renews. The moat holds only if Adobe keeps shipping both AI and non-AI improvements, from Premiere’s rebuilt Color Mode to a 40% reduction in Photoshop crashes, at the pace this base now expects. If the pros feel neglected in favor of casual creators or third-party agents, enterprise creative spend becomes contestable.

Opening the apps to outside agents is still the right call, because buyers are moving toward agents. In Futurum’s 2H 2026 Enterprise Applications Decision Maker Survey (N=833, published August 2026), 42.7% of IT decision-makers prefer AI delivered through autonomous agents, versus 31.2% for copilots. The cost is real, though. When a third-party agent initiates the work, Adobe risks becoming a back-end utility, and utilities get priced like utilities.

Adobe’s Bundled Pricing Faces an Agentic Stress Test

For now, the “every model, one subscription” pitch at Adobe MAX 2026 is working. Creatives get models from Google, Kling, Runway, Flux, OpenAI, ElevenLabs, and Topaz alongside Adobe Fonts, Adobe Stock assets, and storage, without juggling separate accounts and invoices, and Adobe’s tools can auto-select the best model for a task. Buyers are responding: Adobe’s AI-first ending ARR topped $650 million in Q3 FY 2026, up more than 150% year over year.

Predictable seat-based pricing also still has a sizable constituency. In Futurum’s 2H 2026 survey, per-user pricing remains the single most-preferred model among buyers paying for generative AI as an add-on, at 42.3%.

The model gets harder to sustain as agentic workloads become common. Today, most generations are still triggered by a person working in the app. An agent that adapts one poster into dozens of sizes and seven languages, or a Firefly Graph capsule that batches hundreds of product shots, consumes compute in ways that bear no relation to a seat, and third-party agents calling Photoshop through MCP will make that usage even less predictable.

The same survey shows where buyers are heading: 57.7% of add-on buyers prefer consumption- or outcome-based pricing. I expect Adobe to keep the bundled subscription for hands-on creative work while shifting agent-driven and high-volume workloads to metered, per-action, or outcome-linked pricing within 12 to 18 months. Generative credits will be the pressure point, and how transparently Adobe manages that transition will determine whether its loyal professional base sees it as fair or as a stealth price increase.

The Enterprise Story at Adobe MAX 2026 Was Under-Told

MAX is first and foremost a show for creatives, and Adobe was right to center the keynote on the people who live in Photoshop, Premiere, and Illustrator every day. Its enterprise marketing and customer experience story typically gets its full airing at Adobe Summit. Even so, GenStudio received only a few minutes of brand videos from Disney and Coca-Cola at the end of a two-and-a-half-hour keynote, with no live demo, and that undersells an advantage Adobe should press at every event.

CX Enterprise Coworker and Brand Intelligence are arguably the most defensible things Adobe announced, because governing brand consistency across fleets of agents is a problem enterprises will pay to solve, and the Firefly Design demo, in which the agent flagged off-brand fonts and colors across a campaign, hinted at that value.

Enterprise-Safe Positioning Must Extend to Partner Models

Many of the creatives in the MAX audience work inside, or on behalf of, large enterprises, and those organizations care about what Adobe emphasized only in passing: commercially safe Firefly models, security protocols, attribution and ownership of content, and the assurance that Adobe will be around for years to come.

That enterprise-safe positioning matters more, not less, now that third-party models from Google, OpenAI, Kling, and others sit alongside Adobe’s own and outside agents can operate its apps. Enterprise buyers will want clear answers on which models and workflows carry Adobe’s commercial-safety assurances and how governance follows content across agents.

Adobe already has the building blocks. Under Adobe’s Firefly product description, customers on qualifying plans are eligible for IP indemnification for content generated with eligible Firefly features, but that coverage excludes capabilities the interface identifies as powered by non-Adobe-trained models, as well as beta features. Adobe also automatically applies Content Credentials, tamper-evident C2PA metadata that records how an asset was made, to qualifying Firefly-generated content. Those safeguards are anchored in Adobe’s own models.

In an analyst session at MAX, a leader on Adobe’s Firefly Enterprise team said Adobe offers pass-through indemnification on some third-party models, including within its Firefly Creative Production for Enterprise workflow tool, and is working to add more, with its forward-deployed experts helping customers’ legal teams weigh the pros and cons of individual partner models. That is a guided, case-by-case answer rather than a published policy, and one customer in the session, the CIO of event producer Multi Image Group, said indemnification was why his firm chose Adobe in the first place, but that the arrival of partner models made things “a little bit more murky.”

Turning that guidance into clear, published coverage for partner-model and agent-driven output is where Adobe can widen its enterprise lead. Adobe should weave this trust and governance message into its creative narrative between now and Summit, and then prove it at enterprise scale inside Workfront and Adobe Experience Manager.

What to Watch:

• Firefly Builder’s design-to-code pitch pushes Adobe into territory Figma and Canva are already contesting with AI app creation, and private beta status gives both time to respond.
• Google, OpenAI, and other model providers supply Firefly while building their own creative tools; watch whether connectors extend Adobe’s reach or train customers to start their work elsewhere.
• How Adobe meters agent-initiated work and third-party model usage will determine whether the one-subscription promise holds under heavy agent workloads.
• Chakravarthy inherits a creator base anxious about AI and an enterprise business betting on agents, and balancing those two messages will define his first year.
• Disney and Coca-Cola lend credibility, but measurable content supply chain outcomes from GenStudio deployments will matter more to buyers.

You can read the full press release on Adobe’s website.

Declaration of generative AI and AI-assisted technologies in the writing process: This content has been generated with the support of artificial intelligence technologies. Due to the fast pace of content creation and the continuous evolution of data and information, The Futurum Group and its analysts strive to ensure the accuracy and factual integrity of the information presented. However, the opinions and interpretations expressed in this content reflect those of the individual author/analyst. The Futurum Group makes no guarantees regarding the completeness, accuracy, or reliability of any information contained herein. Readers are encouraged to verify facts independently and consult relevant sources for further clarification.

Disclosure: Futurum is a research and advisory firm that engages or has engaged in research, analysis, and advisory services with many technology companies, including those mentioned in this article. The author does not hold any equity positions with any company mentioned in this article.

Analysis and opinions expressed herein are specific to the analyst individually and data and other information that might have been provided for validation, not those of Futurum as a whole.

Read the full Futurum Group Disclosure.

Other Insights From Futurum:

Adobe Brings CX Enterprise Coworker into Third-Party AI Platforms

Documents Are the Overlooked Layer of Enterprise AI Strategy

Adobe and NVIDIA Speed Agentic Creation, but Trust Is What Enterprises Buy

Keith Kirkpatrick is VP & Research Director, Enterprise Software & Digital Workflows for The Futurum Group. Keith has over 25 years of experience in research, marketing, and consulting-based fields.

He has authored in-depth reports and market forecast studies covering artificial intelligence, biometrics, data analytics, robotics, high performance computing, and quantum computing, with a specific focus on the use of these technologies within large enterprise organizations and SMBs. He has also established strong working relationships with the international technology vendor community and is a frequent speaker at industry conferences and events.

In his career as a financial and technology journalist he has written for national and trade publications, including BusinessWeek, CNBC.com, Investment Dealers’ Digest, The Red Herring, The Communications of the ACM, and Mobile Computing & Communications, among others.

He is a member of the Association of Independent Information Professionals (AIIP).

Keith holds dual Bachelor of Arts degrees in Magazine Journalism and Sociology from Syracuse University.

The AI data center boom is looking as shaky as ever.

Nowhere is that more apparent than at the Japanese megafirm SoftBank — perhaps best known for bankrolling WeWork before its spectacular implosion — and which now seems to be standing on a financial snowpack that could collapse into an avalanche.

SoftBank backstopped OpenAI, the lab behind ChatGPT, with over $60 billion in risky debt financing. As the Financial Times reports, that move has now dragged the bank into an ominous loss of $7 billion between July and September, the company’s first quarterly loss in two years of business.

According to the FT, at least $3.7 billion of the loss is likely attributable to a massive OpenAI data center development in Ohio. SoftBank initially blamed that portion on financing for “existing and planned” data center projects there, before backtracking to say the losses are “completely unrelated” to any dealings in the buckeye state.

Compare SoftBank’s $7 billion overall drop to its profit of $2.17 billion the previous quarter, or its heady $15.64 billion profit over the third quarter of 2025, and it’s clear the firm is currently hemorrhaging money to keep some AI dreams alive.

Shares in the company have also plummeted by nearly 50 percent since June, the FT notes, as ratings downgrades and rising skepticism over the viability of the AI industry scare investors into safer corners of the market.

In order to keep pumping money into AI powerlabs like OpenAI, SoftBank itself has started digging into junk bonds — high-risk debt, essentially — raising over $11 billion in non-investment grade bonds in September.

So on top of boasting perhaps one of the highest degrees of exposure to the AI bubble on the planet, the Japanese investment firm is now the world-record holder for the largest corporate junk bond offering ever made. With a foundation this unstable, it’s no wonder SoftBank CEO Masayoshi Son calls the AI bubble a myth — because if it exists, his megabank is in mega trouble.

More on data centers: Only One Town in America Voted in Favor to Building a New Data Center, and the Reason Why Is Bleak

MHL. has taken one of Casio’s most recognisable G-Shock shapes and dialled the styling right back for its latest collaboration.

The new MHL. x G-Shock goes on sale in Japan on 12 November for ¥24,200, through participating MHL. shops and the brand’s online store.

The square watch gets a grey dial with light-grey markings and restrained MHL. branding, set inside a matte resin exterior. G-Shock is hardly a range known for shrinking into the background, but there are no exuberant graphics or conspicuous colour accents here. Understatement is rather the point.

A sizeable premium for the MHL. treatment

The design may be minimalist, but the price premium is less so. Casio currently lists the standard Japanese donor version at ¥14,300, meaning the MHL. edition costs ¥9,900 more, or roughly 69%.

There is plenty of history behind that familiar square shape. Casio dates the first 5600-series model to 1987, four years after the original G-Shock made its debut.

Underneath the quieter wardrobe, the familiar G-Shock credentials remain. The standard model is shock resistant and rated for 200-metre water resistance, while its CR2016 battery is rated for approximately five years. It also offers a stopwatch, 24-hour countdown timer, multifunction alarm and white LED backlight.

The standard watch measures 48.9 × 42.8 × 13.4 mm and weighs 52 g, keeping the familiar compact square proportions intact.

For this collaboration, the changes are primarily aesthetic. MHL. has taken a G-Shock design lineage with nearly four decades behind it and made it look almost conspicuously quiet, which for a G-Shock may be something of an achievement.

Louise is a passionate linguist and performer with a background in translation for the UN and EU institutions. With a deep love for languages and storytelling, she blends precision and creativity, whether translating complex texts or bringing characters to life on stage and screen. When not immersed in words or music, you’ll find her refining accents, working on new songs, or exploring the magic of cinema.