92% of AI Early Adopters Are Seeing A Payoff

Here’s what is in today’s letter:

• Nvidia Announces $500B Investment in U.S. AI Infrastructure

• Kling AI 2.0 Launches with Cinematic Superpowers

• New Global Survey: 92% of AI Early Adopters Are Seeing ROI

• Quick AI Lesson: Create Better AI Image Generation Prompts

• Other AI Tech and News Bytes

Nvidia to Invest $500B in U.S. AI Infrastructure

What’s Going On: Nvidia announced a massive $500B plan to build out a U.S.-based AI infrastructure over the next four years. This includes AI supercomputing hubs and local chip manufacturing facilities designed to protect supply chains against geopolitical risk.

Why It Matters: Nvidia isn't just selling the shovels for the AI gold rush—they’re building the whole mine. This puts the U.S. in a stronger position to compete globally.

Kling AI 2.0 Launches with Cinematic Superpowers

What’s Going On: Chinese AI company Kuaishou has released Kling 2.0, a next-gen text-to-video model that rivals OpenAI’s Sora. Kling can generate realistic, fluid video clips from text prompts—including complex camera angles, physics-aware motion, and talking human avatars. It's one of the most advanced consumer-facing video AIs currently available in China.

Why It Matters: This raises the global stakes in the race for creative AI dominance. Kling’s open release strategy may also pressure U.S. companies like OpenAI and Runway to accelerate their own rollouts.

New Global Survey: 92% of AI Early Adopters Are Seeing ROI

What’s Going On: A new report from Snowflake and Enterprise Strategy Group shows a major shift in how businesses are experiencing AI. After surveying 1,900 business and IT leaders actively using AI, 92% said their AI investments are paying off, and 98% plan to spend more in 2025. On average, early adopters are seeing a 41% return on investment (ROI) — with $1.41 in value for every $1 spent.

Why It Matters: AI isn’t just hype anymore — it’s working. Most companies plan to spend even more on AI in 2025. But there’s a big challenge: Many are struggling to get their data ready for AI.

📌 Fun fact: South Korea leads the world in using open-source AI tools, and the U.S. is best at turning AI into real results.

Quick AI Lesson: Create Better AI Image Generation Prompts

Want to get more impressive, accurate results from image-generation AI? The secret is in your prompt structure. Here's a template that consistently delivers superior results:

Where to use this prompt: You can copy-paste this prompt into popular AI image generators like:

• DALL-E (available through ChatGPT Plus or the OpenAI website)

• Midjourney (on Discord)

• Stable Diffusion (free web versions or desktop app)

• Microsoft Designer (in Microsoft 365)

• Adobe Firefly (in Creative Cloud)

Just replace the bracketed sections with your own ideas!

Copy-paste this prompt framework:

Example:

Beginner tip:

Don't worry about filling in every section at first! Even just completing the subject and setting will give you much better results than a simple description. As you get more comfortable, add more details to refine your images.

Pro tip: Order matters! Start with the subject and its details, then move outward to the setting and atmosphere. This helps the AI prioritize the most important elements of your vision.