[AI DAILY NEWS RUNDOWN] Amazon’s $200B Capex Bet, OpenAI’s Energy Wall, and Meta’s Proprietary Pivot (April 09 2026)

🎧 Listen Ads-Free at https://djamgamind.com

Summary: The latter half of Q2 2026 brings a violent collision between software ambition and physical reality. We perform a forensic analysis of Amazon CEO Andy Jassy’s shareholder letter, defending a $200 billion capex spend while revealing AWS AI revenue has hit a $15 billion run rate. We contrast this with OpenAI pausing its Stargate UK data center due to prohibitive energy costs and regulatory friction. We also deconstruct Meta’s strategic pivot away from open-source with the launch of “Muse Spark,” their new proprietary model from Superintelligence Labs. Finally, we look at Anthropic simplifying the B2B backend with “Managed Agents,” SiFive’s $400M raise to challenge Arm in chip design, and Perplexity hitting a massive $450M ARR.

This episode is made possible by our sponsors:

• DjamgaMind: High-Fidelity Intelligence for the C-Suite. Strategic audio forensics in Enterprise Tech, Healthcare, and Finance. Visit DjamgaMind.com.

Important Topics Covered:

• Andy Jassy’s $200B Defense: Amazon’s 2026 shareholder letter reveals AWS AI revenue at a $15B run rate and custom chips (Trainium/Graviton) crossing $20B.

• The Energy Wall: OpenAI pauses the UK Stargate data center due to exorbitant power costs and regulatory delays, showcasing the physical limits of scaling compute.

• Meta’s Proprietary Pivot: Alexandr Wang’s Superintelligence Labs ships “Muse Spark,” abandoning Meta’s Llama open-source ethos for a closed, monetizable frontier model.

• Anthropic’s Managed Agents: The launch of a $0.08/hr managed backend for enterprise agents, allowing companies like Notion and Rakuten to bypass complex engineering pipelines.

• The Pentagon Blacklist Stands: A federal appeals court upholds the DOD’s blacklisting of Anthropic, citing military security priorities over corporate financial harm.

• Perplexity & SiFive Cash In: Perplexity reaches $450M ARR on usage-based pricing, while SiFive raises $400M at a $3.65B valuation to challenge Arm’s chip design dominance.

Keywords: Andy Jassy shareholder letter, Amazon $200B capex, AWS AI revenue run rate, OpenAI Stargate UK paused, AI energy constraints, Meta Muse Spark, Alexandr Wang Superintelligence Labs, Anthropic Managed Agents, SiFive funding, Perplexity $450M ARR, DjamgaMind, AI Executive Toolkit, AI Unraveled.

🛠️ The AI Executive Toolkit: Stop scrolling through generic lists. Get the hand-picked, forensic-vetted implementation stack to bridge the gap between raw innovation and professional-grade governance. Exclusive listener perks on tools like:

• Chatbase: Build a secure, custom AI chatbot trained exclusively on your firm’s internal data and regulatory documents. (https://link.chatbase.co/etienne-noumen)

• ElevenLabs: Transform lengthy compliance bulletins into high-fidelity “Audio Intelligence” for your team to consume on the go. (https://try.elevenlabs.io/4z7r3skyymar)

• Google Workspace: Professionalize your firm’s infrastructure with secure, cloud-based collaboration and branded communication. (https://referworkspace.app.goo.gl/Q371)

⚗️ PRODUCTION NOTE: We Practice What We Preach.

AI Unraveled is produced using a hybrid “Human-in-the-Loop” workflow.

Meta reenters the AI race with Muse Spark: Superintelligence Labs ships its first model

• Meta has released Muse Spark, the first model from its new Superintelligence Labs division, marking the company’s return to the frontier AI race after a quiet stretch.

• Unlike previous Llama models, Muse Spark isn’t open-weight and can’t be run locally, though Meta says it has plans to open-source future versions of its AI models.

• Independent testing by Artificial Analysis ranked Muse Spark in the top 5 on its Intelligence Index, but the model still trails competitors from OpenAI and Anthropic on agent-based tasks.

• Muse Spark handles voice, text, and image inputs, with a contemplating mode that pits multiple agents against each other on hard problems.

• The model’s benchmarks are competitive with frontier rivals like Opus 4.6 and GPT 5.4 on reasoning, though it lags in coding and tests like ARC-AGI 2.

• Muse Spark is particularly strong in health reasoning, with the company prioritizing the area as part of its ‘personal superintelligence’ mission.

• Unlike the Llama family, Muse Spark is proprietary, with Meta saying it hopes to open-source future versions but has not committed to a timeline.

• Wang took over Meta Superintelligence Labs 9 months ago after Zuck acquired Scale AI for $14.3B, saying the team “rebuilt our AI stack from scratch”.

Why it matters: Meta is back in the game. While still sitting below the top models, Muse Spark is a serious change from where Meta sat with its Llama family. It may not break the internet, but with tons of resources, valuable data across its platforms, and billions of users, Meta’s AI efforts just took a step in the right direction.

Anthropic simplifies the agent-building system

Image source: Anthropic

Anthropic opened a public beta for Claude Managed Agents, a new platform that lets developers go from an agent idea to a live product in days — handling all the backend plumbing that used to take engineering teams months to set up.

The details:

• Users pick the task, tools, and guardrails, with Managed Agents handling running, securing, and controlling what the agentic system can access.

• Agents can work solo for hours without dropping state, with a coordination mode also in preview, letting one agent farm out subtasks to others.

• Notion, Rakuten, Asana, and Sentry are early adopters, with Rakuten reportedly setting up agents across five departments in about a week each.

• Each agent session costs $0.08 per hour on top of the usual AI usage fee, with users paying based on consumption instead of upfront platform fees.

Why it matters: Anthropic continues to roll out features that eat away at the complexities of users getting the most out of their models and tools. Managed Agents now does the same, simplifying the agentic building process and making it possible for anyone to deploy and control agents without the typical backend headaches.

Andy Jassy defends Amazon $200B spending spree

• Amazon CEO Andy Jassy wrote a shareholder letter defending the company’s planned $200 billion in capital spending for 2026, arguing the investments are backed by real customer demand, not guesses.

• Jassy disclosed that AWS’ AI revenue has reached a $15 billion annual run rate, and Amazon’s internal custom chips business is generating over $20 billion a year in value.

• Amazon may sell its Trainium AI chip racks and robotics solutions to outside customers, following the company’s pattern of building tools internally and then offering them as external services.

Tesla is developing a new smaller, cheaper EV

• Tesla is working on a new smaller, cheaper electric SUV that would be a distinct model from the existing Model 3 and Model Y, according to four people familiar with the matter.

• The compact SUV would be about 14 feet long, weigh roughly 1.5 metric tons, use a smaller battery with shorter range, and cost substantially less than the $34,000 entry-level Model 3 in China.

• The project is in early development with production planned for Tesla’s Shanghai factory, though timing is unclear, and the company has a history of starting vehicles that end up delayed or canceled.

Meta removes ads for social media addiction litigation

• Meta started taking down advertisements from lawyers seeking clients who say they were harmed by social media as minors, blocking plaintiff recruitment tied to social media addiction litigation.

• The move comes two weeks after Meta and YouTube were found negligent in a landmark California case, prompting law firms like Morgan & Morgan to run ads on Facebook and Instagram.

• Meta cited its terms of service, saying it will not allow trial lawyers to profit from its platforms while simultaneously claiming they are harmful, and is actively defending against these lawsuits.

Appeals court keeps Pentagon blacklisting of Anthropic in place

• A federal appeals court in Washington, D.C., denied Anthropic’s request to temporarily block the Department of Defense’s blacklisting of the AI company while its lawsuit challenging that decision moves forward.

• The court said the equitable balance favors the government, noting Anthropic faces “relatively contained” financial harm while the DOD is securing AI technology during an active military conflict.

• A separate federal judge in San Francisco last month granted Anthropic a preliminary injunction barring the Trump administration from enforcing a ban on the use of Claude.

OpenAI pauses Stargate UK over energy costs

• OpenAI has paused its Stargate data center project in the UK, pointing to high energy costs and regulatory burdens as the main reasons it cannot commit to long-term infrastructure investment.

• The project, announced last September with Nvidia and Nscale, was tied to the UK’s AI Growth Zone plan, which aimed to create 5,000 jobs and attract £30bn in private investment.

• Stargate’s $500bn US effort is already training AI systems at its Texas facility, with additional projects underway in the UAE and Norway, funded by OpenAI, Oracle, MGX, and SoftBank.

Meta’s internal leaderboard ranks employees by AI token consumption…are we measuring the wrong thing?

A Meta employee built a leaderboard on the company intranet called “Claudeonomics” that ranks token usage across 85k employees. top 250 get ranked. You earn titles like “Token Legend” and “Session Immortal.” 60 trillion tokens burned in 30 days.

It’s not an official thing, but leadership has been pushing hard on AI adoption. But it feels like measuring lines of code written… Volume without outcome tracking is just expensive noise. This cannot be what “good usage” looks like.

Meta pulls ads for anti-Meta law firms:

The Instagram and Facebook owner has been on the losing end of some pretty major legal decisions of late. Most notably, a jury in Los Angeles found the company (along with YouTube) liable for a young woman’s addiction to their products, which exacerbated her mental health struggles. That case set a dangerous precedent for Team Zuck, with law firms now scrambling to develop litigation of their own against the deep-pocketed corporation. Many of these lawyers and firms have been recruiting potential new clients on Meta platforms, of all places. The company put a stop to that this week, removing hundreds of Facebook and IG ads for trial lawyers and marketing companies building class action suits against them. In a statement, Meta explained “we will not allow trial lawyers to profit from our platforms while simultaneously claiming they are harmful.”

Gen Z uses, detests AI:

A new poll from Gallup, ed tech VC firm GSV Ventures, and the Walton Family Foundation (yes, the Walmart Waltons) once more confirms what many other surveys have suggested: Americans are using AI products but aren’t super-happy about it. The number of respondents between 14 and 29 who said they felt “hopeful” about AI was down to 18%, from 27% just a year ago. Nearly a third said that AI makes them feel “angry.” This comes just two weeks after Quinnipiac found that, while 51% of Americans use AI to research topics they’re curious about, 76% said they can trust AI “hardly ever” or only “some of the time.” Quinnipiac found that 35% of Gen Z’ers were “very concerned” about AI’s advancement, while 43% were “somewhat concerned.” These young adults are the exact people AI companies need to adopt their products en masse in order to recoup all of these capital expenditures on energy and compute, so this will be an important trend line to reverse at some point.

SiFive raises $400M:

The AI chip company doesn’t actually manufacture anything. They sell designs and blueprints for new chip concepts to customers like Alphabet/Google. This sector was once largely dominated by the UK’s Arm Holdings, but as they’ve now pivoted over to producing chips of their own — making themselves a rival to many of their best customers — startups like SiFive have spotted a potential opening. The company raised a $400M round at a $3.65 billion valuation. CEO Patrick Little told Reuters he anticipates this will be their last funding round before an IPO.

Andy Jassy Resets AI Narrative

The AI lab horse race continues to volley back and forth every day. Anthropic’s Mythos Preview and Project Glasswing launching on Tuesday, quickly followed by news today that OpenAI also plans to deliver a model with advanced cybersecurity capabilities to key internet infrastructure providers. There’s still debate over how and when these models will roll out to broader audiences. I think this will be an ongoing trend. Cybersecurity is a perfect fit for powerful AI coding agents, and staged releases make sense to allow critical systems to patch vulnerabilities that get exposed by new models. I wouldn’t be surprised to see something similar happen in biosafety, if a powerful model becomes capable of designing a harmful virus, it certainly makes sense to share that with the scientific community through trusted partnerships first, then make sure that capability is carefully under control before releasing a version of the model that can still help you learn about biology broadly.

Andy Jassy zoomed out for his latest 2025 Letter to Shareholders, talking about how AWS followed lots of squiggly lines to get where the company is today.

The original vision included storage, compute, payments, and human intelligence. Some of those (e.g. storage and compute) became lynchpins in AWS. Others didn’t succeed. We didn’t initially plan a database service; and when we built one, our first attempt failed to get traction. We went back to the drawing board and built new relational and non-relational database services, which have resonated well and become core to millions of AWS applications. When we launched EC2 (our compute service), it was a single instance type in one availability zone, Linux-only, with no auto-scaling, load balancing, block storage, or private networking.

Jassy has ramped Amazon capex significantly for good reason, he said: “We’re not investing approximately $200 billion in capex in 2026 on a hunch, AI is a once-in-a-lifetime opportunity where the current growth is unprecedented and the future growth is even bigger.”

He does a great job contextualizing the speed of AI growth by calling back to Thomas Edison. “When Edison opened his first commercial power station in 1882, most people understood it as a better way to light a room. What they couldn’t see was that electricity would eventually reorganize every factory, home, and industry on Earth. AI may have comparable impact. The difference is that electricity took 40 years to get where it was going. AI appears to be moving ten times faster.”

He backs this up by comparing the scale of AI revenue at Amazon to the original launch of AWS. Three years after AWS launched, it had a $58 million revenue run rate. AWS’s current AI revenue run rate is over $15 billion for Q1 2026. Three years after the real start of the AI boom. This is nearly 260 times larger than the AWS growth curve, a testament to the multiplicative power of AI deploying across robust internet and cloud infrastructure.

The AI boom will naturally require a whole lot more infrastructure, and Amazon is “smack in the middle of this land rush” as Jassy puts it. AWS added 3.9 gigawatts of new power capacity in 2025 and expects to double total power capacity by the end of 2027.

Jassy argues (convincingly in my opinion) that every customer experience will be reinvented in the coming years by AI. They aren’t on the frontier of every trend, but have been investing for years in nearly every important category across custom silicon, satellite internet, robotics, same-day delivery, and rural expansion.

What Else Happened in AI on April 09th 2026?

Elon Musk amended his OAI lawsuit to redirect all damages to the nonprofit arm and push Altman off its board, with OAI calling it “a harassment campaign.”

Perplexity hit $450M in estimated annual recurring revenue after a 50% monthly jump, driven by its Computer agentic system and usage-based pricing model.

Elon Musk revealed that xAI has seven new models currently in training on its Colossus 2 supercomputer, including massive 6T and 10T parameter systems.

Canva acquired Simtheory and Ortto, adding agentic AI workspace tools and marketing automation to its platform as it pushes end-to-end campaign workflows.

Jeff Bezos’ secretive AI startup Prometheus poached Kyle Kosic, a former xAI co-founder who led the infrastructure team before leaving the startup for OAI in 2024.

OpenAI published a child safety policy blueprint pushing for updated U.S. laws on AI-generated CSAM, stronger reporting, and built-in safeguards to prevent exploitation.

White-collar workers are quietly rebelling against AI as 80% outright refuse adoption mandates. [LINK]

Microsoft begins removing Copilot from Windows 11, starting with Notepad, Snipping Tool. [LINK]

Reuters: Amazon to stock Lilly’s new weight-loss pill at US kiosks, offer same-day delivery

Reuters: Florida AG opens probe into OpenAI ahead of potential IPO

Bloomberg: Anthropic Completes Tender Offer, But Employees Hold Onto Shares

WSJ: Amazon CEO Presses His Case for Big AI Spending

The Information: Anthropic’s Revenue Growth Suggests OpenAI Is Overvalued

Pirate Wires: Trump to the FAA: Build Me ‘Flying Cars’

Bloomberg: AI-Driven Demand for Gas Turbines Risks a New Energy Crunch

FT: Arm chief Haas in line to lead much of SoftBank’s international business

Nic Carter: One simple reason why I think Satoshi is no longer with us

Aisle: AI Cybersecurity After Mythos: The Jagged Frontier

General Reasoning releases KellyBench, a new long-horizon evaluation for frontier models