Anthropic at $200 billion, Nvidia at $50 billion—will the Fed still raise rates?
Anthropic is preparing for an IPO, and Wall Street has already begun discussing whether it could be worth $200 billion. Nvidia is teaming up with major financial institutions to prepare to introduce more than $500 billion into AI infrastructure. Intel, meanwhile, rode this year’s stock surge and raised $20 billion in market financing in one go.
In the same week, the U.S. CPI, PPI, and consumer data all cooled down together. The market has pushed the probability of a Fed rate hike in September down to around 30%, and gold has returned to around $4,390.
But on the other hand, the auction yields for U.S. 10-year and 30-year long-term Treasuries have risen to levels close to the highest in nearly 20 years and 25 years, respectively.
So this week, we’re mainly watching two things: where the money for AI is coming from, and where the market’s money is heading.
$NVDAB $INTC
Weekly report from August 10 to August 16 is here. If you have more information you’d like to know, feel free to add it—I’ll include it in next week’s report.
NVIDIA did something worth paying more attention to than the release of a new GPU.
It has teamed up with more than 30 technology companies, including Microsoft, IBM, Cloudflare, Palo Alto Networks, Hugging Face, Dell, HPE, and others, to form the Open Secure AI Alliance.
When people hear the word “alliance,” they often think it’s far removed from them. In reality, it’s quite simple: these of the world’s largest tech companies have decided to work together on one question—how to make AI safer?
Why now, specifically? Because AI has started entering the Agent era. In the past, AI only answered your questions. Now, it begins helping businesses write code, operate computers, call APIs, manage databases, and even directly carry out tasks. As AI’s permissions grow larger, security risks are escalating too.
According to a Gartner forecast, by 2028, Fortune 500 companies on average will deploy more than 150,000 AI agents—but today only 13% of companies believe they already have sufficient AI agent governance capabilities. In other words, the pace of AI development has begun to outstrip organizations’ ability to secure it.
And on top of that, the recent security incident at Hugging Face has caused the whole industry to rethink what truly needs to be defended against in the future: not only hackers, but hackers using AI to launch attacks—possibly even using AI agents to infiltrate enterprise systems.
So instead of each company defending on its own, it’s better to jointly develop security tools, share vulnerability information, and establish security standards so the entire industry can improve its defensive capabilities together. That’s also why NVIDIA spearheaded the formation of this alliance.
I think what’s truly worth watching about this alliance isn’t which companies have joined—it’s the important signal it sends.