Subscribe
Share
Share
Embed
Anthropic just hit $30B annualized revenue while its two most powerful models are suspended by a US export control order. This episode breaks down what that says about enterprise AI adoption, plus Meta's 1.6-gigawatt compute bet and why the power gri
Beyond Brief Daily is your signal in the noise.
Every day, we cut through the scroll so you don't have to. This is where curiosity meets clarity — quick hits on the trends, tools, and stories actually worth your time.
What you'll hear: AI breakthroughs reshaping how we work and build. Business and brand moves that matter for founders and creators. Real tools — no hype, just what's working right now. And the weird, wonderful stuff the internet can't stop talking about — from UFOs to culture wars to rabbit holes you didn't know you needed.
New episodes daily. Stay curious. Stay ahead.
Beyond Brief Daily — I'm Michael Benatar. AI, tech, business. Let's get into it.
Anthropic just overtook OpenAI in revenue — and the timing is almost absurd. Anthropic is running at $30 billion annualized. OpenAI, which raised at an $852 billion valuation just months ago, is behind them. The crossover happened while Anthropic's two most powerful models — Claude Fable 5 and Mythos 5 — are sitting offline, suspended by a US Commerce Department export control directive that dropped June 12. The government's argument is that the models pose a national security risk. Anthropic's argument is that any jailbreak the government found is already replicable with other publicly available tools. What's bizarre is who's locked out: Samsung and SK Hynix — the companies supplying the memory chips that American AI runs on — can't access Claude's best models because Anthropic can't verify nationality in real time. Today is also the refund deadline for customers who bought usage credits for integrations that are now dark.
The revenue story and the ban story are the same story — Anthropic is the most commercially dangerous AI company right now, and the government just hit pause on its best products. Wrong. The ban doesn't slow Anthropic down. It proves how embedded they already are.
On the infrastructure side, Meta signed a 1.6-gigawatt compute deal with startup Crusoe — two data centers in Texas and Missouri. One company. One point six gigawatts. And Meta is just one player. The five biggest hyperscalers are collectively on pace to spend over $700 billion in capital expenditures this year. Revenue is growing fast but not fast enough to justify that spend at current multiples. If model improvement slows even modestly, there's a lot of stranded infrastructure coming. Do the math.
I go deeper on the AI infrastructure buildout every week in the newsletter — theBeyondbrief.com if you want the full picture.
FERC moved this week to address part of that problem. The Federal Energy Regulatory Commission ordered six major grid operators to fast-track interconnection requests from data centers — the first system-wide federal intervention to clear AI's power bottleneck. Grid operators now have 30 days to report available generating capacity. Data centers foot the bill for connection costs. But the order doesn't create new power — it just speeds up access to what's already there. And in Q1 alone, grassroots groups blocked 75 data center projects worth $130 billion, organizing around electricity costs, water, and noise. Fast-lane grid access doesn't solve a generation shortage. Watch this one — the power constraint is more binding than the model constraint right now.
SpaceX is now a $2 trillion company. The IPO — largest in stock market history — opened at $135 a share and kept going. xAI sits inside that umbrella. Private ambition and public markets are fully merged now, and the valuation debate around AI just got a lot louder.
Quick hit before the close: 10% of people globally now use AI chatbots for news every week, according to the Reuters Institute's 2026 Digital News Report. Only 4% click through to original source articles. Publishers built paywalls to protect their content. AI read it anyway, summarized it, and now holds the audience. That line just got crossed — and there's no obvious way back for publishers.
What connects all of this: the AI industry is hitting friction from every direction at once — regulatory blocks on models, power grid constraints on infrastructure, audience capture eating media. Anthropic's revenue milestone, achieved with its best models offline, tells you how deep enterprise adoption already runs. At Benatar Brands, we're advising clients to diversify their model dependencies now, before the next export directive or grid delay makes that decision for them. Single-model stacks are a liability.
That's your brief. Follow the show on Instagram @thebeyondbrief, find me on X @MichaelBenatar, and if you want this in your inbox every morning — theBeyondbrief.com. I'm Michael Benatar. See you tomorrow.