Podbit · Moonshots with Peter Diamandis
KIMI K3's published architecture contains no revolutionary breakthrough — it's still essentially a transformer with well-understood improvements in mixture of experts and linearized attention. The terrifying implication: if you can nearly match GPT-5.5 Max with a recognizable transformer, what are the American frontier labs spending billions of dollars on?
Nvidia disclosed a 9.3% passive stake in Nebius worth more than $4 billion — 22.3 million Class A shares. The neo-cloud company's stock surged 18.8% on the news, following a $2 billion strategic investment Nvidia made in March.
Reports say Oracle could be required to post roughly $7 billion in collateral tied to a large AI data center project in Wisconsin. The stock is already trading just above its 52-week low, with investors increasingly worried about Oracle's balance sheet and infrastructure spending.
Treasury Secretary Scott Bessant has warned the U.S. could sanction Chinese AI companies for improperly using American AI models. The warning comes after Anthropic told the Senate that Alibaba executed what it called the largest known distillation attack against it to date.
AI investment has been dominated by infrastructure spending, but Armani sees the next wave coming in software applications, particularly healthcare and cybersecurity. TSMC's $265 billion expansion underscores the infrastructure boom that's now priming the application layer.
Despite the alarm over models like KIMI K3 and Qwen, Armani argues there's no second Sputnik moment — Chinese AI uses the same transformer architecture as American AI. The race is global, but not as existentially asymmetric as the headlines suggest.
AI can clone a person's voice with alarming ease, undermining voice-based banking authentication systems. This isn't a future risk — Armani warns it's already a practical threat that financial institutions need to address now.
Ransomware groups have become so sophisticated they now offer customer service to their victims. Armani highlights this as a sign of how organized and dangerous the cybercrime ecosystem has become — and why AI-powered defense is urgent.
Palo Alto Networks is at the forefront of AI-driven cybersecurity defense, while Anthropic represents the responsible AI development model. Armani sees compliance and caution as non-negotiable in an era where AI can be weaponized.
AI's brand collapse isn't Sam Altman's fault — it's Washington's. When governments don't regulate, don't protect jobs, and let billionaires crowd the White House, ordinary people conclude the game is rigged — and they're right.
We use essential and analytics cookies to run Vuci. To understand how the site is used: Privacy Policy.
Install Vuci on your phone
Add it to your home screen for a faster, app-like experience.
Install Vuci on your phone
Tap the Share button, then “Add to Home Screen”.
A new version is available
Reload to get the latest Vuci.