It’s Thursday
Here’s what’s worth your time today.
Nvidia CEO Jensen Huang said the company is nearing completion of its planned investment in OpenAI, part of a deal announced last year that could reach up to $100 billion. The agreement has taken months to finalize, raising questions about whether it might stall, but Huang told investors the partnership is progressing and remains a priority. The move reflects Nvidia’s strategy to stay deeply embedded across the entire AI ecosystem, even as it also supplies chips and infrastructure to rivals like Anthropic, Google, and xAI. The signal is structural: Nvidia isn’t betting on one model provider — it’s positioning itself as the backbone of the entire AI industry.
Supported by EnergyX
Thanks to growing demand across high-growth sectors like AI and robotics, lithium stock prices grew 2X+ from June 2025 to January 2026.
$ALB climbed as high as 227%. $LAC hit 151%. $SQM, 159%. But the real winner may be a stock not listed on public exchanges, EnergyX.
This $1B unicorn’s patented technology can recover up to 3X more lithium than traditional methods, earning investment from leaders like General Motors. Now they’re preparing for commercial production just as experts project 5X demand growth by 2040.
They’ve announced what could be one of the US’ largest lithium production facilities and have rights to ~150,000 lithium-rich acres across the Americas.
TLDR: Companies used to brag about AI pilots. Now, saying you’re “still piloting” signals you’re behind.
Mentioning AI pilots on earnings calls fell 18% in late 2025, as investors and executives increasingly see pilots as experiments that never ship. A widely cited study found 95% of enterprise AI pilots failed to deliver measurable financial impact, fueling terms like “pilot purgatory.” Companies like AT&T and Bristol-Myers Squibb are avoiding the label entirely, emphasizing production deployments and measurable value instead. The signal is clear: the market is shifting from AI experimentation to execution, and investors now reward real deployment, not proofs of concept.
Chinese AI lab DeepSeek has withheld its upcoming V4 model from U.S. chipmakers like Nvidia and AMD, instead giving early access to domestic partners including Huawei. That breaks with standard practice, where AI labs work closely with chipmakers to optimize performance ahead of release. The move could help Chinese hardware catch up and reflects growing geopolitical tension around AI infrastructure. The signal is strategic: models and chips are now intertwined assets, and China is increasingly building its stack independently of U.S. suppliers.
Nvidia reported $68B in quarterly revenue, up 73% year over year, with $62B coming from data centers alone. CEO Jensen Huang said demand is so high that even older GPUs are fully utilized. The company is also close to finalizing a major partnership with OpenAI. The signal is clear: AI infrastructure spending is still accelerating, and compute has become the foundation of the entire AI economy.
Two years ago, Anthropic was a niche startup. Today, its Claude models are reshaping national security policy, financial markets, and how startups are built. The company is now valued at $380 billion, and engineers consistently rank Claude among the best models for reasoning, coding, and reliability. That lead is already forcing reactions across the industry. OpenAI is accelerating new releases to close the gap, while Chinese labs like DeepSeek are preparing competing models that could reset the balance again.
Claude’s influence is showing up far beyond Silicon Valley. The Pentagon is pressuring Anthropic to loosen military guardrails because Claude is already embedded in classified systems and considered difficult to replace. On Wall Street, new Claude releases have triggered sharp sell-offs in legal, cybersecurity, and financial software stocks, reflecting fears that AI agents could replace entire SaaS categories. And among startups, tools like Claude Code are dramatically compressing development timelines, letting small teams build products that once required much larger engineering orgs.
The broader shift is becoming clear. Frontier models are no longer just tools. They’re becoming infrastructure that shapes geopolitics, markets, and the structure of software itself.