Hey,
Let’s unpack what’s new…
TOP 3
SpaceX’s long-awaited IPO filing finally dropped, but the numbers surprised a lot of people. Despite targeting a staggering $1.75 trillion valuation, the filing showed SpaceX lost nearly $5 billion last year on under $19 billion in revenue, with Starlink standing out as its main profitable business. Even Musk’s AI division, which includes xAI and X, generated far less revenue than many expected.
At almost the exact same time, OpenAI is reportedly preparing its own confidential IPO filing, setting up what could become the defining public market battle of the AI era. Both companies are now racing toward public markets with enormous valuations tied less to current fundamentals and more to future dominance in AI, infrastructure, and compute.
The bigger signal is that AI IPOs are no longer being priced like traditional tech companies. Investors are increasingly betting on who controls the future layers of intelligence, compute, and distribution, even if the underlying economics still look messy today.
Supported by Sonar
When speed is prioritized over reliability, technical debt and security risks accumulate silently. To realize the full ROI of AI, organizations must shift from manual reviews to an automated verification layer.
This guide helps you make that shift — and choose the right partner to support it:
Six evaluation pillars for modern code review.
Strategies to solve the AI productivity paradox.
A comprehensive checklist for code health.
TOP 3
Anthropic reportedly told investors it expects to more than double quarterly revenue to roughly $10.9 billion and post its first-ever operating profit, marking a major milestone in the AI race. The profitability may not last all year because of massive compute spending ahead, but the signal still matters.
For most of the past two years, the AI industry has operated under the assumption that frontier labs would burn money indefinitely while chasing scale. Anthropic now looks like one of the first major labs proving there may actually be a path to building a profitable AI business before going public.
The timing is also interesting. The news landed the same day reports surfaced that OpenAI is preparing its IPO filing, further sharpening the contrast between the two rivals. OpenAI still dominates consumer mindshare, but Anthropic is increasingly positioning itself as the enterprise-focused company with stronger economics, faster business adoption, and tighter operational discipline.
TOP 3
Nvidia crushed earnings again, raised its dividend from 1 cent to 25 cents per share, and approved another $80 billion in buybacks. Normally, that’s the kind of move mature tech giants make once growth starts slowing and they run out of obvious places to aggressively reinvest capital.
But Jensen Huang used the earnings call to pitch something bigger instead: a completely new $200 billion market built around AI agents.
The idea revolves around Nvidia’s new Vera CPU, which Huang says is specifically designed for “agentic AI.” While GPUs handle the heavy model reasoning, Nvidia believes future AI agents will rely heavily on CPUs to actually run tasks, use tools, and operate autonomously across software systems. Huang’s bet is that the world won’t just have billions of AI users. It’ll eventually have billions of AI agents constantly working in the background.
The bigger signal is that Nvidia is trying to expand beyond simply powering AI models. It wants to own the infrastructure layer underneath the entire agent economy too. And at the same time, Wall Street is quietly asking a different question: when a company starts handing this much cash back to shareholders, does it mean the hypergrowth phase is slowly starting to mature?
In other developments
Former Calm CEO David Ko says the AI era will require people to actively protect their critical thinking, use time saved by AI more intentionally, and rethink how children interact with the technology. He argues AI is evolving even faster than social media and warns companies should ask whether they’d be comfortable with their own children using the products they build. (Axios)
xAI plans to spend another $2.8 billion on gas turbines for its AI infrastructure, including the same “mobile” generators at the center of an environmental lawsuit over pollution near its Memphis data center. The move highlights the growing tension between the AI industry’s massive power demands and the environmental impact of rapidly scaling compute infrastructure. (TechCrunch)
An OpenAI reasoning model has reportedly solved a major open problem in discrete geometry that mathematicians had studied for nearly 80 years, marking one of the strongest examples yet of AI contributing original mathematical research. The breakthrough is being described by researchers as a milestone for both AI reasoning and human-AI collaboration in science. (OpenAI)
Advertise with us
Looking to promote your company, product, service, or event to 100,000+ AI leaders?