Tag: Anthropic

🎧 Listen Ads-Free: Tired of interruptions? Subscribe to AI Unraveled directly on Apple Podcasts

Important Topics Covered:

• Anthropic’s B2B Dominance: Hitting a $30B ARR, with over 1,000 businesses spending >$1M annually, and locking in multiple gigawatts of Google Cloud TPU capacity by 2027.

• Project Glasswing: Anthropic giving unreleased ‘Mythos’ access to Broadcom, Cisco, CrowdStrike, and JPMorgan to harden critical cybersecurity infrastructure.

• The OpenAI Expose & Social Contract: Deconstructing the New Yorker piece featuring Ilya Sutskever’s memos, alongside Altman’s push for a 4-day workweek and sovereign AI wealth fund.

• Intel Joins Terafab: The $25B semiconductor complex in Texas partnering Intel, Tesla, SpaceX, and xAI to produce 1 terawatt of compute capacity annually.

• Nvidia’s 5-Layer Infrastructure Cake: Why Nvidia argues AI is the new electricity, segmented into Energy, Chips, Infrastructure, Models, and Applications.

• Meta’s “Tokenmaxxing”: Inside the bizarre “Claudeonomics” leaderboard where Meta employees burned 60 trillion tokens in 30 days without proving ROI.

• Corporate Liability: Target formally shifting the legal burden of AI shopping assistant errors onto its customers.

This episode is made possible by our sponsors:

🛑 AIRIA: Secure your AI workforce. AIRIA unifies orchestration, security, and governance into a single command center, using micro-VM sandboxing to protect sensitive data from agentic goal-hijacking. 👉 Govern your agents: [LINK]

🎙 DjamgaMind: High-Fidelity Intelligence for the C-Suite. If you are a modern decision-maker, DjamgaMind delivers strategic audio forensics in Healthcare, Energy, and Finance. Stop reading headlines and start understanding the systemic impact with our human-verified, technical-grade analysis. 👉 Explore the Forensics: https://DjamgaMind.com/regulations

🛠️ 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)

• Fireflies.ai: Automate your meeting forensics by transcribing and extracting action items from your clinical consultations and board meetings. (https://fireflies.ai/?fpr=etienne62)

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

Full toolkit at: https://djamgamind.com/toolkit

⚗️ PRODUCTION NOTE: We Practice What We Preach.

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

$30B ARR shows Anthropic’s strategy is working

While rivals zigged, Anthropic zagged, and the bet keeps paying off.

On Monday, the company announced that its revenue run-rate, or its financial forecast based on current performance, has surpassed $30 billion. This figure is nearly triple the run rate at the end of 2025, which came in at $9 billion, and double that of mid-February, which the company reported was $14 billion. By the end of February, CEO Dario Amodei confirmed annual revenue had exceeded $19 billion.

This exponential growth results from an equally notable increase in customer demand. In February, Anthropic disclosed that over 500 business customers were spending over $1 million, and now reports that the number exceeds 1,000 businesses, representing a two-fold growth in less than two months.

The rapid growth can be mostly attributed to Anthropic’s laser focus on enterprise. For instance, Claude Code, a go-to coding tool for many developers, alone generated a run-rate revenue of over $2.5 billion by February.

In that same month, weekly active users have also more than doubled since January 1, and business subscriptions to Claude Code have quadrupled since the beginning of 2026. Those figures are likely much larger now. This success has allowed the company to bridge the gap with the much bigger OpenAI, despite the company kick-starting the AI race as we know it.

At the end of February, OpenAI topped $25 billion in annualized revenue, according to a report from The Information, citing a person familiar with the figure, representing a 17% increase from the annualized revenue it generated at the end of the year. The comparison isn’t perfectly apples-to-apples, as the two companies use different methods to calculate revenue, as noted by the WSJ. But with an IPO in the works, the ChatGPT maker has recently made moves to pivot towards enterprise, just like Anthropic.

Last month, it shut down its Sora generative AI video platform, ended its $1 billion content partnership with Disney, and put its ChatGPT “adult mode” on hold, all while OpenAI CTO Sarah Friar acknowledged that enterprise “is a very profitable business at scale” and that it’s how OpenAI will “build a sustainable business model.”

To meet growing demand and scale up the computing it needs to power it, Anthropic also announced an expansion of its existing partnership with Google Cloud and Broadcom by signing a new agreement. The resulting multiple-gigawatt TPU capacity is expected to come online beginning in 2027, according to the blog post.

OpenAI’s new ‘social contract’ ideas for society, ASI

Image source: OpenAI

OpenAI just published a 13-page policy document with ideas to help society navigate superintelligence and its societal impacts, asking Washington to tax AI-driven profits, create a wealth fund, implement a 4-day workweek, and more.

The details:

• The proposal said we are “beginning a transition toward superintelligence”, with Altman telling Axios the moment requires a new “social contract” for society.

• The most aggressive idea: a sovereign-style fund seeded by AI firms that would pay dividends to every American, as Alaska does with oil revenue.

• Other ideas include taxes on robot labor, a 4-day workweek, “Right to AI” access for all, and containment playbooks for rogue autonomous AI.

• Axios called it “the most detailed blueprint any tech titan has ever published for how to tax, regulate, and redistribute wealth from the technology he’s building.”

Why it matters: The CEO of an $852B company is asking the U.S. to prepare for a future where his own tech breaks the economic system — and you don’t make that pitch unless you believe it’s actually coming. But with the way things are moving, coupled with the slow-moving gears of the government, the clock is ticking.

New Yorker surfaces memos behind Altman’s firing

Image source: The New Yorker

The New Yorker published an investigation into Sam Altman, drawing on 100+ interviews, unseen memos from ex–chief scientist Ilya Sutskever, and notes from Dario Amodei — alleging a long-running pattern of deception at the top of OpenAI.

The details:

• The reporting spans Altman’s full career arc, including conflicts at his startup Loopt, Y Combinator partners trying to push him out, and the OAI board drama.

• Sutskever’s memos, built from 70 pages of Slack messages and HR docs, allege Altman misrepresented safety protocols to the board.

• Amodei’s private notes, kept for years, reach the same conclusion as Sutskever independently: “The problem with OpenAI is Sam himself.”

• A Microsoft exec told the reporters there’s “a small but real chance” Altman is “remembered as a Bernie Madoff, Sam Bankman-Fried-level scammer.”

Why it matters: While there is no ‘smoking gun’ in this piece, there is a vastly detailed and concerning pattern of deception that seems to span across Altman’s career. But for every detractor, you also have fiercely loyal supporters and coworkers – making the CEO of the nearly trillion-dollar AI giant one of the most polarizing figures in the world.

Meta to open-source new AI models

• Meta plans to release open-source versions of its next-generation AI models, which are derived from two proprietary frontier models codenamed Avocado and Mango expected to launch this year.

• The open-source versions won’t include all features found in the closed-source editions, possibly lacking certain neural networks, having smaller parameter counts, or skipping post-training steps.

• AI safety is reportedly one reason Meta will hold back features, and the company does not expect its upcoming models to beat competitors like Anthropic and OpenAI across the board.

Intel joins Elon Musk’s $25B Terafab AI chip project

• Intel has officially joined Elon Musk’s Terafab project, a $20–25 billion semiconductor complex planned for Austin, Texas, partnering alongside Tesla, SpaceX, and xAI to build chips at scale.

• The facility aims to produce 1 terawatt per year of compute capacity by manufacturing edge-inference processors for Tesla’s FSD systems and radiation-hardened chips for SpaceX satellites and xAI.

• Intel CEO Lip-Bu Tan hosted Musk at Intel facilities before the announcement, and the company will contribute its process technology, high-volume fabrication, and packaging expertise to the project.

Anthropic doubles down on Google Cloud TPUs

• Anthropic announced an expanded partnership with Google Cloud, securing access to multiple gigawatts of TPU capacity to train and run its AI models starting in 2027.

• The deal delivers Google’s Tensor Processing Units through Google Cloud infrastructure with hardware from Broadcom, giving Anthropic enormous compute for its Claude family of AI systems.

• Anthropic is also adopting Google Cloud tools like BigQuery, Cloud Run, and AlloyDB, while thousands of companies already access Claude models through Google Cloud today.

Meta employees compete on internal AI usage leaderboard

• Meta has an internal leaderboard called “Claudeonomics” where employees compete to consume the most AI tokens, tracking usage across more than 85,000 workers on the company intranet.

• Employees burned through 60 trillion tokens in just 30 days, with the top user averaging 281 billion, though some simply leave AI agents running for hours to pad their numbers.

• Despite Silicon Valley treating “tokenmaxxing” as a productivity metric, nobody has put up hard numbers proving that high token consumption actually translates into real business results or revenue gains.

Anthropic is giving some firms access to Claude Mythos to bolster cybersecurity defenses

Anthropic is giving a group of Big Tech and cybersecurity firms access to a preview version of Claude Mythos—its unreleased and most advanced AI model—in an attempt to bolster cybersecurity defenses across some of the world’s most critical systems.

The company has been concerned that the new model may pose unprecedented cybersecurity risks and increase the likelihood of large-scale AI-driven cyberattacks this year.

The initiative, called Project Glasswing, allows companies, including Amazon Web Services, Apple, Broadcom, Cisco, CrowdStrike, Google, JPMorganChase, Microsoft, and Nvidia, to use Anthropic’s Mythos Preview for defensive security work and share their learnings with the wider industry. Anthropic is also providing access to roughly 40 more organizations responsible for building or maintaining critical software infrastructure, allowing them to use the model to scan and secure both their own systems and open-source code.

Read more: https://fortune.com/2026/04/07/anthropic-claude-mythos-model-project-glasswing-cybersecurity/

Why AI may be more like electricity than software

If you think AI is the world’s next smartphone or cloud, Nvidia would like to change your mind.

Instead, Nvidia thinks AI equates more to underlying infrastructure like electricity or the internet. At HumanX in San Francisco on Monday, this was the topic of the opening keynote of the event, which brings together 6,000 people from across the AI industry to discuss how to use AI to solve business problems.

I’d tweak this idea to say that today’s AI will evolve into intelligence so ubiquitous that we’ll rarely mention it. But it’s likely to power the next generation of tools, as well as the next set of technological and scientific breakthroughs.

Those promises are why the AI industry is in the midst of “the largest infrastructure buildout in human history,” as HumanX CEO Stefan Weitz called it during Monday’s keynote.

That’s because all that intelligence requires a lot of hardware to run, and it’s going to require a lot more in the years ahead as the number of people using AI continues to grow and those who are using it today keep finding more things to do with it.

“The connection between compute and intelligence is stronger than ever,” Bryan Catanzaro, VP of applied deep learning research at Nvidia, said on Monday during the opening panel.

Nvidia has extrapolated all the technology it takes to power AI by coining the phrase “AI is a five-layer cake,” which is the same phrase HumanX used for the opening keynote on Monday. Here’s how the layers break down:

1. Energy: Everything starts with power, and right now, this is the greatest constraint to scaling up AI to meet future demand.

2. Chips: Today’s AI workloads need GPUs to run tasks in parallel at a massive scale, high-bandwidth memory to move data at breakneck speeds, and fast interconnections between all the pieces.

3. Infrastructure: Here’s where the physical components come together, from land to construction to power delivery to networking to cooling to server racks. This is where AI factories are emerging.

4. Models: The AI models understand topics across a ton of different domains, and now that’s expanding to other kinds of models as well, from scientific discoveries to autonomous systems to robotics.

Applications: The place where almost all of the value is created remains the application layer. It’s where we get daily tools, agents, coding helpers, self-driving cars, industrial robots, and lots of other things being invented day by day.

What Else Happened in AI on April 07th 2026?

The Information: Meta Employees Vie for AI ‘Token Legend’ Status

Anthropic expands partnership with Google and Broadcom for multiple gigawatts of next-generation compute

The Deep View: Anthropic passes $30B ARR after Pentagon clash

Hermeus raises $350M at a $1B valuation

Reuters: Intel joins Musk’s Terafab AI chip project to power humanoid, data center goals

WSJ: What to Know About OpenAI’s Ideas for a World With ‘Superintelligence’

Bloomberg: OpenAI, Anthropic, Google Unite to Combat Model Copying in China

WSJ: Inside a Corporate Retreat That Went Very Badly Wrong

NYT: Bill Ackman’s Fund Offers to Buy Universal Music in $64 Billion Deal

NYT: Shots Fired at Indianapolis Councilman’s Home, After Vote Backing Data Center

AP says it will offer buyouts as part of pivot away from newspaper-focused history

Wired: Edward ‘Big Balls’ Coristine Is Helping Out on Viral Fraud Videos Now

The Hollywood Reporter: Kalshi Forecasts Coming to Fox News as Prediction Market Strikes Deal With Cable News Leader

Bloomberg: Cerebras Backer Eclipse Raises $1.3 Billion for Robotics, AI Infrastructure

WSJ: Five Amazing Tech Innovations We Should Expect in the Next 25 Years

The Wrap: McClatchy Journalists Revolt Against AI: ‘It’s a Betrayal’ | Exclusive

The Information reported that Sam Altman and OpenAI CFO Sarah Friar are not on the same page regarding IPO timing, though OpenAI denied the claim.

Iran’s military singled out the $30B Stargate data center in Abu Dhabi as a target, publishing satellite footage and vowing to destroy U.S. infrastructure across the region.

OpenAI Head of Business Finance Chengpeng Mou posted new stats on ChatGPT’s use for healthcare questions, with the platform getting 2M insurance messages weekly.

Google released AI Edge Eloquent, a free iOS dictation app that cleans up raw speech into polished text entirely on-device.

Legion Health won approval to let its AI app directly refill psychiatric medications, the first time a state has greenlit AI to do the process without clinician oversight.

Target puts customers on the hook for AI shopping assistant errors [Link]

🎧 Listen Ads-Free: Tired of interruptions? Subscribe to AI Unraveled directly on Apple Podcasts at

Summary: The first Monday of Q2 2026 reveals massive fractures in the capital structure of the AI industry. We perform a forensic analysis of the internal conflict at OpenAI, where CFO Sarah Friar has been sidelined after projecting a catastrophic $200 billion cash burn, directly challenging CEO Sam Altman’s aggressive IPO timeline. We also deconstruct Altman’s controversial new policy blueprint calling for a tax on automated corporate labor. At the infrastructure layer, we analyze Nvidia’s brilliant strategic hedge: releasing frontier-class open models (Nemotron 3) to prevent OpenAI and Anthropic from monopolizing the software ecosystem and dictating hardware prices. Finally, we look at Anthropic’s margin squeeze, forcing third-party agents off flat-rate billing, and the dangerous rise of “Cognitive Surrender” in enterprise decision-making.

Important Topics Covered:

• The OpenAI IPO Civil War

• The “Automated Labor Tax”

• Nvidia’s Strategic Hedge

• The Death of Flat-Rate Compute

• Cognitive Surrender:

• LinkedIn’s “BrowserGate”

• Netflix VOID

Keywords: OpenAI IPO readiness, Sarah Friar, Sam Altman AI Tax, OpenAI $200B cash burn, Nvidia Nemotron 3 open source, AI hardware monopoly, Anthropic Claude OpenClaw ban, Penn study cognitive surrender, LinkedIn BrowserGate, Netflix VOID physics-aware AI, DjamgaMind, AIRIA.

This episode is made possible by our sponsors:

🛑 AIRIA: Secure your AI workforce. AIRIA unifies orchestration, security, and governance into a single command center, using micro-VM sandboxing to protect sensitive data from agentic goal-hijacking. 👉 Govern your agents: [LINK]

🎙 DjamgaMind: High-Fidelity Intelligence for the C-Suite. If you are a modern decision-maker, DjamgaMind delivers strategic audio forensics in Healthcare, Energy, and Finance. Stop reading headlines and start understanding the systemic impact with our human-verified, technical-grade analysis. 👉 Explore the Forensics: https://DjamgaMind.com/regulations

🛠️ 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)

• Fireflies.ai: Automate your meeting forensics by transcribing and extracting action items from your clinical consultations and board meetings. (https://fireflies.ai/?fpr=etienne62)

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

Full toolkit at: https://djamgamind.com/toolkit

⚗️ PRODUCTION NOTE: We Practice What We Preach.

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

Sam Altman proposes AI tax and regulation blueprint

• OpenAI CEO Sam Altman released a 13-page policy blueprint on Monday that proposes new taxes, a public wealth fund, and regulation to prepare for AI’s expected impact on jobs and the economy.

• The document calls for taxes “related to automated labor” to protect funding for programs like Social Security and SNAP, and recommends giving every citizen a stake in AI-driven economic growth.

• Altman also suggested employers and unions push for four-day workweeks with no pay cuts, expanded training for human-centered jobs, and guardrails on how the government can deploy AI systems.

Why Nvidia chose open models to reshape AI

If you’re wondering why AI chips leader Nvidia is now building open models that compete with the Chinese open-source champs, and even proprietary models from OpenAI and Anthropic, then you’re not alone.

Last month, Nvidia launched Nemotron 3 Super, a 120-billion-parameter reasoning model that outperformed expectations in benchmarks. This is a mixture-of-experts model with a 1-million-token context window. In other words, it’s a serious model made to compete with the frontier labs. Meanwhile, the company promised that a model 4x its size, to be called Nemotron 3 Ultra, is coming soon.

And because Nvidia opens the weights, datasets, and training recipes, it’s among the most open models in the world, especially for a model of this capability. Some of the only models that could claim to be more open would be the ones from MBZUAI, which The Deep View covered in depth in January. But Nvidia’s open models are far closer to full-stack openness than most of the open-source models, which only offer open-weight releases.

So why would the leading hardware company of the AI era make software that competes with its leading customers?

“We’re not trying to control AI. We’re trying to grow it,” Bryan Catanzaro, VP of applied deep learning research at Nvidia, told The Deep View. “And so our incentives as a company, our business is aligned with open models and with supporting the ecosystem in a very direct way.”

Kari Briski, VP of generative AI software at Nvidia, told The Deep View another perspective: “The model is the byproduct. It is not core to our business, which allows us to just open up the data, open up the recipes, open up everything.”

If we break it down, there are three benefits Nvidia gets from making its own models:

1. Extreme hardware co-design: Making their own models allows Nvidia to optimize the heck out of their GPUs, CPUs and other hardware to run AI. They don’t have to wait to get the latest models from the frontier labs to plan the next stage of optimizations.

2. Hedging against proprietary monopolies: If the frontier labs that need the latest and greatest hardware dwindle down to only a handful of players, then Nvidia could end up at their mercy. When you have a smaller number of customers you rely on for huge numbers of orders, then those customers gain more and more control over your prices. They can demand lower prices because they know so much of your business depends on them.

3. Letting a thousands flowers (a.k.a. customers) bloom: By releasing open models that other hardware and software makers can use as a rapid on-ramp to build their own AI products and serve the various niches in the industry, Nvidia is powering up the ecosystem, helping companies with limited resources have models they can use to compete and potentially creating a lot more future customers when those companies succeed and grow.

“You don’t want one person winning [because] then they decide all the rules. You need a big open ecosystem for everybody to come along,” said Briski.

Anthropic boots third-party agents from Claude plans

Anthropic just blocked agent platforms like OpenClaw from running on Claude plans, requiring users to pay separately via usage add-ons or API keys, as the company confronts agent-driven demand its flat-rate pricing was never built to absorb.

The details:

• Agent tools hit Claude with nonstop requests that exceed what its normal plans typically cover, despite Anthropic models being the leading driver for the tech.

• Anthropic’s Boris Cherny announced the change, saying it is a step towards “managing growth to continue to serve our customers sustainably long-term”.

• Anthropic is handing out credits worth a month’s subscription, discounting add-ons up to 30%, and offering refunds amid cancellation requests.

• OpenClaw creator Peter Steinberger criticized the step, saying, “First they copy popular features into their closed harness, then they lock out open source.”

Why it matters: Anthropic was already catching heat over tighter rate limits, and walling off its agentic power-user community won’t help the goodwill problem. It’s a tough situation with Anthropic’s agent usage likely playing a role in degrading normal user experience, but OAI is now there as the alternative at a crucial time in the rivalry.

When AI thinks, humans stop questioning

AI might be causing us to forget how to think for ourselves.

Recent research from the University of Pennsylvania found that AI users were often willing to accept flawed AI reasoning, readily incorporating it into their decision-making with “minimal friction or skepticism.”

The research documents the rise of “cognitive surrender,” a phenomenon in which users adopt AI outputs while “overriding intuition… and deliberation.”

• In a study of nearly 1,400 participants across 9,500 trials, researchers found that subjects accepted unsound AI reasoning more than 73% of the time and only overruled models’ decisions about 20% of the time.

• Additionally, participants with higher trust in AI and “lower need for cognition and fluid intelligence” tended to fall victim to this more often.

“Across domains, AI tools are not merely assisting decision-making; they are becoming decision-makers,” The research reads. “This shift opens new theoretical ground: How should we understand human cognition and decision-making in an age when we outsource thinking to artificial processes?”

The study adds to a growing body of research on how AI may be impacting the way that we think. One of the most commonly cited studies comes from the MIT Media Lab, in which a group of test subjects was asked to write SAT questions with three different tools: one with OpenAI’s ChatGPT, one with Google search, and one with no help at all. Consistently, the ChatGPT users “underperformed at neural, linguistic, and behavioral levels.”

Even some of AI’s biggest names are questioning its effects on our brains. Anthropic CEO Dario Amodei said in a March interview with podcaster Nikhil Kamath that deploying AI in the wrong ways could easily make people “become stupider,” but only if they choose to forgo learning entirely. “Even if an AI is always going to be better than you at something, you can still learn that thing. You can still enrich yourself intellectually,” Amodei told Kamath.

The researchers, however, posit that cognitive surrender may not inherently be a bad thing. If an AI model is generally better at reasoning and decision-making than the person using it, with fewer mistakes, “deferring to a statistically superior system may be adaptive or even optimal.”

The bigger issue, however, comes down to agency. The researchers noted that this trend could mark a profound shift in cognition itself, “one in which users may not know when or why they have deferred, and where the line between human and machine agency becomes blurred

Netflix opens physics-aware AI for video editing

Image source: Netflix Research

Netflix just released VOID, an open-source framework built to erase video objects while rewriting the physics associated with them, instead of typical erasing and inpainting tools.

The details:

• Existing removal tools just paint over backgrounds, without actually reasoning about the cause-and-effect those edits introduce across the broader scene.

• VOID uses a mask that maps what to erase, what’s physically affected, and what to keep, with a judge model then charting the consequences.

• VOID can handles physics it never trained on, with demos like a balloon floating when a holder is removed or blocks not falling when one in the chain is erased.

• 25 evaluators compared VOID against six baseline models including Runway, preferring Netflix’s results nearly 2/3 of the time.

Why it matters: This is Netflix Research’s first public AI release, and its a sign of where the video space is heading — intuitive systems that don’t just erase objects in footage like an image editor, but can actually simulate and alter the physics of the scene based on the changes for more controllability and real production use.

LinkedIn secretly scans over 6000 browser extensions

• A report called BrowserGate accuses LinkedIn of running hidden code that scans for over 6,000 browser extensions on users’ computers, linking specific software choices back to real people and their employers.

• The investigation claims LinkedIn can infer personal details like religious beliefs, political views, or job-seeking activity, and also scans for over 200 competing products like Lusha, Apollo, and ZoomInfo.

• LinkedIn denies the accusations, saying it checks for extensions only to stop scammers and scraping, while the report’s author is a developer whose account was restricted for breaking platform rules.

China forces Apple to remove Jack Dorsey’s Bitchat

• Apple pulled Jack Dorsey’s decentralized messaging app Bitchat from its China App Store after Beijing’s internet regulator, the Cyberspace Administration of China, said it violated rules on services capable of social mobilization.

• Bitchat works entirely over Bluetooth and mesh networks without internet connectivity, letting messages hop between devices — a design that has made it popular during government-imposed connectivity blackouts in multiple countries.

• The app has passed three million total downloads across platforms, and this is the second time China has targeted a Dorsey-backed decentralized app, after banning the Nostr-based Damus in 2023.

OpenAI CFO questions readiness for 2026 IPO

• OpenAI’s CFO Sarah Friar has told colleagues the company is not ready for an initial public offering by late 2026, putting her in direct conflict with CEO Sam Altman’s goal to list by Q4.

• Internal projections show OpenAI burning through more than $200 billion before reaching positive cash flow, with losses for 2026 alone projected at roughly $14 billion against $2 billion in monthly revenue.

• Friar no longer reports to Altman directly, has been excluded from key financial meetings, and the company has quietly retained Goldman Sachs and Morgan Stanley to manage a possible offering.

Vibe coding boosted App Store submissions in 2025

• App Store submissions surged 84 percent year-over-year in Q1 2026, and the growth of vibe coding tools like Claude Code and ChatGPT Codex is believed to be driving the increase.

• For the full year of 2025, submissions grew 30 percent versus 2024, nearly hitting 600,000 total, with momentum building each quarter and accelerating sharply into early 2026.

• Apple says its review team processes 90 percent of submissions within 48 hours, but developers and consumers have complained about lower-quality apps flooding the App Store as a result.

What Else happened in AI on April 06th 2026?

OpenAI is navigating a leadership change, with Fidji Simo on medical leave, COO Brad Lightcap on special projects, and CMO Kate Rouch stepping down for cancer recovery.

Anthropic acquired startup Coefficient Bio for roughly $400M, folding the team into its healthcare and life sciences group focused on drug discovery.

Mercor confirmed a data breach tied to an attack on open-source library LiteLLM, with hackers claiming access to up to 4 TB of data from the $10B AI training startup.

Pika Labs released PikaStream 1.0 in beta, a real-time model that lets AI agents join Google Meet calls as video avatars with voice cloning and live conversation.

OpenAI rolled out ChatGPT in CarPlay, allowing users to access Voice Mode in their supported vehicle for hands-free use.

MIT study models AI ‘sycophancy’, warns of ‘delusional spiraling’ in chatbot interactions [Link]

18-month New Yorker investigation finds OpenAI’s Sam Altman lobbied against the same AI regulations he publicly advocated for, pursued billions from Gulf autocracies, and how he tried to hide a post-firing investigation that produced no written report [Link]

Japan Wants to Build a Solar Ring Around the Moon That Will Provide Endless Clean Energy to Earth [Link]

UK confirms drone-killing DragonFire laser weapon for Royal Navy destroyers by 2027 —laser downs 400mph high‑speed drones, costs $13 per shot [Link]

Meta salary data reveals a VP of AI can make $650,000 in base salary [Link]

Microsoft says Copilot is for entertainment purposes only, not serious use — firm pushing AI hard to consumers and businesses tells users not to rely on it for important advice [Link]

OpenAI proposes superintelligence governance plan – taxing automation, establishing AI wealth funds, 4-day work weeks [Link]

🎧 Listen Ads-Free: Tired of interruptions? Subscribe to AI Unraveled directly on Apple Podcasts

or at https://djamgamind.com

Summary: The first week of April 2026 brought a violent collision between infinite software scaling and finite physical reality. We perform a forensic autopsy on OpenAI’s historic $122 billion raise and the internal executive reshuffle happening right before their IPO. We analyze Elon Musk’s unprecedented move to force SpaceX IPO underwriters to buy Grok AI subscriptions, and the chilling supply chain data revealing that 50% of US data centers are paralyzed by transformer shortages. Finally, we look at the human impact: how a solo founder used everyday AI tools to build a $1.8 billion company, while everyday adults retreat from the social internet in record numbers.

This episode is made possible by our sponsors

🎙 DjamgaMind: High-Fidelity Intelligence for the C-Suite. If you are a modern decision-maker, DjamgaMind delivers strategic audio forensics in Healthcare, Energy, and Finance. Stop reading headlines and start understanding the systemic impact with our human-verified, technical-grade analysis. 👉 Explore the Forensics: https://DjamgaMind.com/regulations

🛠️ 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, ElevenLabs, AIRIA, and Google Workspace. 👉 Get the Toolkit: https://DjamgaMind.com/Toolkit

Important Topics Covered:

• OpenAI’s Trillion-Dollar Trajectory: The $122B raise ($852B valuation), leadership reshuffles (Lightcap, Simo), and the acquisition of the TBPN tech show.

• The Secondary Market Pivot: Why investors are abandoning $600M in OpenAI stock to deploy $2B into Anthropic.

• The SpaceX IPO Leverage: Elon Musk forces banks to subscribe to xAI’s Grok to participate in the historic $1.75T SpaceX public offering.

• The Physical Compute Wall: Data shows 50% of US data centers are delayed. China still accounts for over 40% of U.S. battery imports and 30% of key switchgear components.

• The $1.8B Solo Unicorn: Matthew Gallagher scales Medvi to $401 million in Year 1 (and a projected $1.8B this year) with just $20,000 and AI agents.

• The Social Retreat: A new Ofcom report reveals UK adults posting on social media dropped from 61% to 49%, while ChatGPT usage surged from 31% to 54%.

• Ecosystem Wars: Anthropic blocks OpenClaw, Apple preps an iOS 27 Siri App Store, and Microsoft injects Copilot ads into GitHub pull requests.

Keywords: OpenAI $122B funding round, SpaceX $1.75T IPO, Elon Musk Grok subscriptions, US data center delays, Chinese battery imports AI, Matthew Gallagher Medvi, Solo Unicorn startup, Anthropic Claude Code leak, Microsoft Copilot GitHub ads, Meta Mercor data breach, Google Gemma 4, DjamgaMind, AIRIA, AI Executive Toolkit, AI Unraveled.

⚗️ PRODUCTION NOTE: We Practice What We Preach.

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

OpenAI reshuffles leadership as top executives step back

• OpenAI is making several leadership changes at once, with its COO moving to a new role, its CEO of AGI development taking medical leave, and its marketing head stepping down.

• COO Brad Lightcap will now lead “special projects” involving complex deals and investments, while former Slack CEO Denise Dresser, the new chief revenue officer, takes over his commercial duties.

• Fidji Simo is taking medical leave for several weeks to deal with a neuroimmune condition, and co-founder Greg Brockman will manage product while she is away.

Musk requires SpaceX IPO banks to buy Grok subscriptions

• Elon Musk is requiring banks, law firms, and other advisers working on the SpaceX IPO to buy subscriptions to Grok, his AI chatbot, according to a New York Times report.

• Some banks have agreed to spend tens of millions on Grok and have already started integrating the chatbot into their IT systems as part of the deal.

• The IPO filing came after SpaceX purchased xAI, which makes Grok and owns the X social network, while Grok faces investigations for generating harmful imagery.

NASA astronauts took iPhones on Artemis II mission

• NASA astronauts on the Artemis II mission carried iPhone 17 Pro Max devices to the moon, marking the first time the agency gave each crew member a personal smartphone for photos and videos.

• The iPhones can’t connect to the internet or use Bluetooth and are limited to capturing photos and videos, and NASA had to run a four-phase safety review covering hazards like shattering glass.

• The crew also has four GoPro Hero 11 cameras and two Nikon D5 bodies onboard, and NASA discussed using Velcro to mount phones inside the Orion capsule in microgravity conditions.

Claude just shut the door on OpenClaw

• Anthropic announced that starting April 4, 2026, Claude Pro and Max subscribers will no longer be able to connect their subscriptions to third-party agentic tools like OpenClaw, citing strain on compute and engineering resources.

• Users can still power third-party agents with Claude models, but they must now switch to pay-as-you-go “extra usage” billing or Anthropic’s API, which charges per token instead of offering flat-rate access.

• OpenClaw creator Peter Steinberger, now at OpenAI, accused Anthropic of copying popular open-source features into Claude Code and then locking out competing tools, while Anthropic offered credits and discounts to ease the transition.

Meta pauses Mercor partnership after AI data breach

• Meta has paused its partnership with Mercor, the AI training data company, after a massive data breach exposed as much as 4TB of sensitive information including candidate profiles and personally identifiable information.

• Mercor has been approaching professionals across industries, including visual effects artists, offering payment in exchange for work materials from previous jobs like “4D physics scenes with camera data.”

• Much of the material Mercor is seeking likely belongs to former employers and is protected by intellectual property laws and confidentiality agreements, even though the company says it “does not buy intellectual property.”

Half of planned US data center builds delayed or canceled

• About half of all planned U.S. data center builds this year are expected to be delayed or canceled, mainly because key electrical equipment like transformers, switchgear, and batteries is in short supply.

• Lead times for high-power transformers in the U.S. have stretched from around two years before 2020 to as long as five years today, far exceeding the under-18-month deployment cycles AI data centers need.

• China still accounts for over 40% of U.S. battery imports and nearly 30% of certain transformer and switchgear categories, so ongoing trade-war tensions could further disrupt builds despite massive spending plans.

OpenAI acquires tech talk show TBPN

• OpenAI has bought TBPN, a popular daily tech talk show on YouTube and X, marking the AI company’s first acquisition of a media property.

• TBPN, hosted by John Coogan and Jordi Hays, will keep its own brand and editorial independence but report to OpenAI’s chief political operative, Chris Lehane.

• The deal raises conflict-of-interest questions since OpenAI, a company approaching an IPO, now owns a show that regularly covers OpenAI itself and its competitors.

NASA sends astronauts toward the Moon for first time in 50 years

• NASA launched four astronauts toward the Moon on Wednesday aboard the Space Launch System and Orion capsule, marking the first crewed flight beyond low Earth orbit in more than 50 years.

• Pilot Victor Glover will manually fly Orion within about 30 feet of the spent upper stage in a rendezvous demonstration, testing all six degrees of freedom to build confidence for future lunar dockings.

• NASA calls Artemis II a test flight with off-ramps, meaning controllers will check life-support performance after launch vibration and bring the crew home early if systems do not meet expectations.

Google rethinks the AI model race with Gemma 4

• Google launched Gemma 4, its next generation of open models, positioning the smaller and more efficient family as a strong alternative as inference costs rise across the industry.

• Gemma 4 comes in four sizes — from 2B-parameter models that run offline on phones and laptops to a 31B Dense model ranked #3 on the Arena AI text leaderboard.

• The open-source models now feature an Apache 2.0 license, and Google says the Gemma family has been downloaded over 400 million times since its first launch in February 2024.

AI helped two brothers build a $1.8B company

• Matthew Gallagher used more than a dozen A.I. tools and just $20,000 to build Medvi, a telehealth GLP-1 drug provider that hit $401 million in sales its first year with only one employee — his brother.

• Gallagher relied on ChatGPT, Claude, Grok, Midjourney and Runway to write code, generate ads and handle customer service, while CareValidate and OpenLoop managed doctors, pharmacies and shipping.

• Medvi is now on track for $1.8 billion in sales this year and has expanded into men’s health and meal delivery, with a 16.2 percent net profit margin — triple that of competitor Hims.

Investors flee OpenAI for rival Anthropic LINK

• Investors on secondary markets are turning away from OpenAI shares and rushing to buy equity in rival Anthropic, with some large OpenAI stakes now nearly impossible to sell.

• About $600 million in OpenAI shares from institutional investors found no buyers, while secondary platforms report over $2 billion in cash ready to deploy into Anthropic.

• Investors see better risk-reward in Anthropic at its $380 billion valuation, betting it will close the gap with OpenAI’s $852 billion, especially given Anthropic’s stronger enterprise client growth.

Microsoft launches 3 new AI models to rival OpenAI

• Microsoft released three in-house AI models — MAI-Transcribe-1, MAI-Voice-1, and MAI-Image-2 — covering speech-to-text, voice generation, and image creation, competing directly with OpenAI, Google, and ElevenLabs.

• Mustafa Suleyman told VentureBeat that teams of fewer than 10 engineers built the audio and image models, and MAI-Transcribe-1 runs on half the GPUs of competitors while beating Whisper on all 25 benchmarked languages.

• Suleyman confirmed Microsoft plans to build a frontier large language model and become “completely independent,” following a renegotiated OpenAI contract that now lets Microsoft pursue superintelligence on its own.

Cloudflare launches WordPress competitor

• Cloudflare has launched EmDash, an open source CMS it calls the “spiritual successor” to WordPress, designed to be more secure and built on what the company describes as an “AI native” architecture.

• EmDash runs each plugin in an isolated sandbox called Dynamic Workers, requiring plugins to declare permissions upfront, since Cloudflare says 96% of WordPress vulnerabilities come from plugins with unrestricted access.

• The new CMS is built on a scale-to-zero principle that only bills for CPU time during actual requests, and WordPress users can migrate by importing a WXR file or installing the EmDash Exporter plugin.

Amazon in talks to acquire Globalstar for $9 billion

• Amazon is in talks to buy Globalstar, a satellite communications company valued at around $8.81 billion, as it tries to grow its early-stage Leo satellite internet service, the Financial Times reported.

• Apple’s 20% stake in Globalstar, part of a $1.5 billion investment in 2024 to expand satellite and ground infrastructure, has complicated the deal and required separate negotiations between Amazon and Apple.

• Amazon’s Leo program has about 200 satellites in orbit and plans for 7,700, but it still trails SpaceX’s Starlink, which operates over 9,600 satellites and serves more than nine million users.

Fewer adults are posting on social media

• A new Ofcom report found that fewer adults in the UK are posting, sharing, or commenting on social media, dropping from 61% in 2024 to 49% as platforms shift toward video.

• Nearly half of adults are now concerned about historic posts causing problems later in life, with worries about professional prospects and reputation driving people to stop posting permanently.

• Meanwhile, active use of AI tools like ChatGPT has jumped from 31% to 54% among UK adults, and fewer social media users believe the apps are good for their mental health.

Alibaba launches 3 closed-source AI models in 3 days

• Alibaba released three closed-source AI models in three days this week, ending with Qwen3.6-Plus, a coding and multimodal reasoning model sold through paid APIs to enterprise customers.

• The shift follows the departure of Qwen’s technical lead Lin Junyang in early March, with one contributor suggesting the exit was not voluntary, and Alibaba replacing him with a Google DeepMind veteran.

• Alibaba is targeting $100 billion in cloud revenue within five years, and Qwen3.6-Plus scores 78.8 on SWE-bench Verified, trailing only Claude Opus 4.5 among the models it compared against.

SpaceX confidentially files for biggest IPO in history

• SpaceX has confidentially filed IPO paperwork with the SEC, reportedly seeking a $1.75 trillion valuation in what would be the largest initial public offering in history at $75 billion raised.

• The company lined up 21 banks to manage the offering, internally codenamed “Project Apex,” dwarfing Saudi Aramco’s $29 billion listing in 2019, which was previously the biggest IPO ever.

• SpaceX needs the money to build its Starship rocket, replenish Starlink satellites, and pay for compute powering xAI’s deep learning models after absorbing Musk’s AI lab in February.

Anthropic accidentally leaks Claude Code

• Anthropic accidentally exposed over 512,000 lines of Claude Code source code when someone made a packaging error while pushing out version 2.1.88 of the software on Tuesday.

• The leak revealed the full software scaffolding around Claude Code — instructions telling the model how to behave, what tools to use, and where its limits are — not the AI model itself.

• This marks the second accidental exposure in a week, after Anthropic made nearly 3,000 internal files publicly available last Thursday, including a draft blog post describing an unannounced model.

Oracle cuts thousands of jobs to boost AI spending

• Oracle laid off thousands of employees across multiple countries this week, with analysts estimating cuts could reach 30,000 positions to free up billions in cash flow for AI data center construction.

• Workers received termination emails from “Oracle Leadership” at 6 a.m. with no prior warning, after being locked out of internal systems at three in the morning Pacific time.

• The cuts come as Oracle spends $50 billion on capital expenditure this year and carries a $156 billion total buildout estimate, even as its stock has lost nearly half its value since September.

OpenAI raises record $122B at $852B valuation

• OpenAI closed a $122 billion funding round at an $852 billion valuation, making it the largest private fundraising event in history, with Amazon leading at $35 billion and individual investors buying shares through ARK ETFs.

• Amazon’s $35 billion tranche carries conditions tied to an IPO milestone, effectively making a public listing a requirement of OpenAI’s own financing rather than a voluntary strategic decision.

• At $852 billion, OpenAI’s private valuation now exceeds every publicly listed tech company except Apple, Microsoft, Nvidia, and Alphabet, setting up a tense gap between pre-IPO price and market reality.

Google prepares a screenless Fitbit band to rival Whoop

• Google is working on a screenless Fitbit fitness band designed to compete with Whoop and Oura, combining simple hardware with AI-powered health coaching and a subscription-based model for extra features.

• The band is described as a grey fabric design with an orange inner lining, and it will rely on a redesigned Fitbit app featuring an AI personal health coach covering mental wellbeing, cycle tracking, nutrition, and hydration.

• Basketball player Stephen Curry shared a sponsored video teasing the device, and Google confirmed he has been collaborating with its team, with more details and a full launch expected later this year.

Baidu robotaxis freeze in Wuhan causing traffic chaos

• More than 100 Baidu robotaxis stopped running in Wuhan due to a system malfunction, stranding passengers in fast-moving traffic on a ring road, according to police and Chinese media reports.

• Some passengers were afraid to exit because their vehicle had stopped in the middle lane with other cars passing on both sides, while others pushed an SOS button and left on their own.

• This is the first reported mass shutdown of robotaxis in China, and Baidu, which operates more than 1,000 driverless taxis mostly in China, did not have any immediate comment.

Meta tests paid subscriptions on Instagram

• Meta is testing a paid subscription called Instagram Plus in a few countries, giving everyday users access to exclusive features for a small monthly fee.

• Subscribers can view Stories without the poster knowing, see how many people rewatched their Stories, create unlimited audience lists, and extend or spotlight their Stories.

• Pricing varies by country — roughly $1 to $2 per month in Mexico, Japan, and the Philippines — and the subscription is separate from Meta Verified, which targets creators and businesses.

Meta launches prescription AI glasses at $499

• Meta and EssilorLuxottica launched two new Ray-Ban Meta smart glasses styles, the Blayzer Optics and Scriber Optics, starting at $499 and designed to work with prescription lenses including progressive and transition options.

• Both frames feature slimmer designs, swappable nosepads, and adjustable temple tips for a better fit, and they are available for pre-order now with sales starting April 14 at retailers like LensCrafters.

• Meta is also rolling out new software features for all Ray-Ban Meta glasses, including Japanese, Mandarin, and Arabic translation support, food and nutrition tracking, and AI-powered message thread summaries.

Google now lets US users change their Gmail address

• Google is now rolling out a feature that lets users in the U.S. change their Gmail address without creating a new account or losing access to their existing data.

• Users can only change their username once every 12 months, and their old email address will be preserved as an alternate address that still works for signing in.

• The feature is rolling out gradually, so not everyone will see the “Change Google Account email” button in their Personal info settings right away, according to Google’s support page.

OpenAI’s Sora burned $1 million daily before shutdown

• OpenAI’s video-generation tool Sora was losing roughly a million dollars per day before the company shut it down last week, just six months after its public launch.

• Sora’s user count peaked at around one million but then dropped below 500,000, while video generation kept consuming expensive AI chips at a rate OpenAI could not justify.

• Disney had committed $1 billion to a partnership with OpenAI around Sora but learned about the shutdown less than an hour before the public announcement, killing the deal.

Apple may launch a dedicated Siri AI app store

• Apple is reportedly planning to let users install third-party AI chatbots inside Siri through a new feature called “Extensions,” with a dedicated section in the App Store acting as an AI marketplace.

• Bloomberg’s Mark Gurman reports that the changes are coming in iOS 27, which should get its first developer beta in June around Apple’s Worldwide Developers Conference starting June 8.

• Apple already integrated OpenAI’s ChatGPT into Siri, but Extensions will offer a wider choice of AI agents that users can browse and install from the new App Store section.

Microsoft Copilot now shows ads in pull requests

• Microsoft’s Copilot AI tool has started inserting promotional messages into pull requests on GitHub, advertising a Raycast extension for the Copilot coding agent on macOS and Windows machines.

• A search on GitHub shows the exact same promotional text appearing in over 11,000 different pull requests across thousands of repos, and even merge requests on GitLab contain the injection.

• Hidden HTML comments labeled “START COPILOT CODING AGENT TIPS” in the raw markdown suggest Microsoft is inserting these ads, likely to promote its own developer ecosystem or partner integrations.

ChatGPT app store struggles six months post-launch

• OpenAI’s push to turn ChatGPT into an app platform with over 300 integrations from companies like Booking and StubHub is off to a sluggish start six months after launch.

• Partner companies are hesitant to hand off customer relationships and payments to OpenAI, so most apps force users to leave ChatGPT to complete purchases or even view basic details.

• Developers have complained about a tedious app-approval process, buggy development tools, and a lack of usage data, leaving them “running quite blind” on whether their apps are working properly.

Eli Lilly bets $2.75 billion on AI drug discovery

• Eli Lilly announced a $2.75 billion deal with Hong Kong-based Insilico Medicine to use artificial intelligence for drug discovery and development, giving Lilly exclusive license to manufacture and market the resulting therapies.

• Insilico CEO Alex Zhavoronkov told CNBC his company has used AI to develop at least 28 drugs, with close to half already at a clinical stage of testing.

• Eli Lilly has also committed $1 billion over five years with Nvidia to finance talent, infrastructure and computing needed to address bottlenecks in AI-based drug discovery.