Are you looking for easy breakdowns and knowledge deep dives on the latest key concepts to help you understand business and finance topics? Finance news doesn’t have to be confusing. 1440’s weekly newsletter explains markets, policy, and investing in plain English—complete with charts, context, and links to learn more.

Knowledge Starts Here

⚛️ Meta's nuclear power deal with Constellation Energy exemplifies how AI's voracious energy appetite is breathing new life into America's aging nuclear plants. The 20-year agreement will expand Illinois' Clinton Clean Energy Center by 30 megawatts starting in 2027—strategically timed to begin just as state subsidies expire, ensuring the facility's survival without taxpayer support. This follows similar moves by Microsoft (reviving Three Mile Island), Amazon (investing in small modular reactors), and Google (backing advanced nuclear projects), as tech giants scramble to secure reliable, carbon-free baseload power for their 24/7 AI operations. The rush toward nuclear partnerships reveals a stark reality: AI's computational demands are fundamentally reshaping both the tech and energy sectors, with 25 states passing nuclear-friendly legislation last year to attract these lucrative partnerships.

✏️ MIT and Stanford's SketchAgent represents a fundamental shift in AI art generation—moving from creating polished images to mimicking the messy, iterative process of human sketching. The system, powered by Anthropic's Claude 3.5 Sonnet, generates drawings stroke-by-stroke from text prompts, allowing users to either watch AI sketch autonomously or collaborate in real-time, much like working with a human drawing partner. By focusing on sketching as a thinking tool rather than an artistic endpoint, researchers have tapped into something profound: the way humans naturally use quick visual representations to workshop ideas, from DNA helices on whiteboards to napkin diagrams of business concepts. This approach could democratize visual communication, making it easier for anyone—regardless of artistic skill—to translate abstract thoughts into concrete visual forms that AI can understand and build upon.

🧊 LG Uplus's liquid cooling test facility signals a critical inflection point for AI infrastructure—the heat generated by modern AI workloads has finally overwhelmed traditional air cooling's capabilities. The South Korean telecom giant is evaluating two radical approaches at its Anyang facility: direct-to-chip cooling that circulates coolant through plates attached to processors, and full immersion systems that literally bathe servers in thermally conductive oil. This isn't just about efficiency metrics; it's about enabling the next generation of AI computing, as traditional cooling methods simply cannot handle the thermal density of modern GPU clusters. With partners like Vertiv and CoolIT Systems, LG Uplus is betting that mastering liquid cooling will be as crucial to AI competitiveness as securing compute power itself—a recognition that sustainable infrastructure may determine who can afford to scale AI operations in the coming years.

📊 ComScore's new AI tracking data reveals that artificial intelligence has quietly crossed a critical threshold—30% of Americans now use AI tools monthly, with mobile devices driving this silent revolution. The analytics giant is now monitoring 117 AI services across nine categories, finding that 67 million Americans access these tools via smartphones, where creative applications dominate: audio AI leads with 23.8 million users, followed closely by image generation and design tools at 23 million each. While ChatGPT reigns supreme across all platforms, the surprise emergence of creative tools as category leaders suggests Americans are using AI less as a novelty chatbot and more as a practical creative assistant. This shift from experimental technology to mainstream digital behavior marks a fundamental change in how we need to think about AI adoption—it's no longer about whether people will use AI, but which tools they'll choose for their daily creative and productivity needs.

🇨🇭 Microsoft's $400 million Swiss investment reveals how the AI infrastructure battle is increasingly being fought on regional terms, with tech giants building local capabilities to navigate Europe's complex regulatory landscape. The expansion of four data centers near Zurich and Geneva specifically targets Switzerland's heavily regulated healthcare, finance, and government sectors—markets that require data residency and compliance guarantees American cloud services alone cannot provide. Beyond hardware, Microsoft's commitment to train one million Swiss residents in AI skills by 2027 acknowledges a critical reality: infrastructure without expertise creates dependency, not development. With 80% of Swiss business leaders viewing 2025 as pivotal for AI strategy and Azure OpenAI usage jumping to 31%, this investment positions Switzerland as a crucial testing ground for how American tech companies can successfully embed themselves in Europe's distinct AI ecosystem while respecting local sovereignty concerns.

🚗 Day.ai's vision to become "the Waymo of CRM" reveals a fascinating paradox in enterprise AI adoption: users desperately want automation but refuse to take their hands off the wheel. Founded by former HubSpot executives Christopher O'Donnell and Michael Pici, the startup discovered that their most valuable features weren't sophisticated pipeline automation but simple daily touches—three priority tasks each morning and meeting talking points that save five minutes of prep time. Their "fingers on the wheel" approach mirrors autonomous vehicles' gradual rollout, acknowledging that trust in AI systems handling critical business relationships must be earned incrementally. Perhaps most intriguingly, O'Donnell frames their ambition through video game graphics evolution, comparing legacy CRMs to 8-bit Mario while positioning Day.ai as the photorealistic Elden Ring—suggesting they're not just building better software but reimagining what a CRM should feel like in the AI era.

📱Microsoft's integration of Sora into Bing mobile demonstrates how quickly cutting-edge AI can become a commodity feature—just 10 months after Sora's splashy debut, it's now a free tool buried in a search app menu. The implementation reveals both opportunity and constraint: while anyone with a smartphone can now generate AI videos from text prompts, the five-second vertical format limit suggests we're witnessing AI's awkward adolescence, where revolutionary capabilities meet mobile computing realities. Microsoft's tiered system—10 fast generations before throttling to slower speeds—cleverly gamifies the experience while managing computational costs, even allowing users to trade Microsoft Rewards points for more instant gratification. Yet the most telling detail may be that Sora, once the undisputed leader in AI video, now needs Bing's distribution muscle to compete with rivals like Runway and Luma that have caught up or surpassed its capabilities, proving that in the AI race, yesterday's breakthrough is today's table stakes.