- Evolving AI Insights
- Posts
- 🔋 U.S. vs. China race may already be over
🔋 U.S. vs. China race may already be over
Also: GPT-5 backlash and Altman’s trillion-dollar plan

Welcome, AI enthusiasts
AI experts return from China stunned: the U.S. power grid is so weak the race may already be over. In China, energy is seen as a solved problem, while in the U.S. it has become the bottleneck that could shape the future of AI. Let’s dive in!
In today’s insights:
U.S. grid struggles as China powers ahead in AI
GPT-5 backlash and Altman’s trillion-dollar plan
Coding dreams collide with AI reality
Read time: 4 minutes
LATEST DEVELOPMENTS
AI & ENERGY
🔋 U.S. Grid Struggles as China Powers Ahead in AI
Evolving AI: Experts returning from China warn the U.S. grid may not be able to support AI’s rapid growth.
Key Points:
China treats electricity as a solved problem, while U.S. grids face critical strain.
AI’s power demand could add $6.7T in global data center costs by 2030.
U.S. infrastructure delays contrast with China’s long-term overbuilding strategy.
Details:
AI experts visiting China say the country’s energy system is years ahead of the U.S. when it comes to supporting AI growth. “Everywhere we went, people treated energy availability as a given,” Rui Ma wrote on X after returning from a recent tour of China’s AI hubs. In the U.S., Goldman Sachs says AI’s surging power demand is already straining the grid. China’s massive overcapacity means data centers are seen as a way to absorb excess supply, backed by reserve margins that stay between 80% - 100%. By contrast, U.S. grids often operate around 15%, leaving little room to handle sudden spikes. That gap is reinforced by China’s long-term, state-led energy planning, while the U.S. relies on short-term private capital and faces delays, regulation, and grid instability.
Why It Matters:
The real game-changer isn’t the race to fix power, it’s that China already has. While U.S. data centers could take up 6.7 to 12 percent of electricity by 2028, driving load pressure and higher bills, China treats AI hubs as a natural outlet for its long-term energy surplus. That cushion lets China scale AI at full speed while the U.S. struggles just to expand capacity. And with a clear race between the U.S. and China, this is a major advantage that could decide who leads the next wave of AI.
Typing is a thing of the past.
Typeless turns your raw, unfiltered voice into beautifully polished writing - in real time.
It works like magic, feels like cheating, and allows your thoughts to flow more freely than ever before.
Your voice is your strength. Typeless turns it into a superpower.
Evolving AI: GPT-5’s launch has drawn sharp criticism, but Sam Altman says OpenAI is preparing to spend “trillions” on infrastructure.
Key Points:
GPT-5’s release sparked ridicule, with users calling it worse than earlier models.
Sam Altman admitted rollout mistakes but noted API traffic doubled within 48 hours.
Despite calling AI a bubble, Altman says OpenAI will spend trillions on data centers.
Details:
OpenAI’s newest model, GPT-5, has faced heavy backlash since launch, with many users calling it underwhelming. At a dinner with reporters in San Francisco, CEO Sam Altman acknowledged the rollout was mishandled but highlighted growth: API traffic doubled, GPU shortages continue, and ChatGPT reached record usage. Altman also admitted the AI sector may be in a “bubble” but compared it to the dot-com boom, where hype surrounded real progress. He told reporters OpenAI expects to invest trillions in building data centers to power future AI systems, underscoring both the vast ambitions and financial risks tied to scaling.
Why It Matters:
GPT-5 may be getting mocked, but usage is climbing and OpenAI is doubling down with plans to spend trillions on infrastructure. It shows how AI can be criticized as underwhelming yet still become so embedded that scale, not polish, drives the next phase.
COMPUTER SCIENCE GRADUATES
💻 Coding Dreams Collide with AI Reality
Evolving AI: New computer science graduates face rising unemployment as AI tools reshape the entry-level job market.
Key Points:
Students were promised six-figure tech salaries but now face unemployment or jobs outside tech.
AI coding assistants reduce demand for junior engineers, hitting new grads hardest.
Some turn to non-tech roles, side hustles, or AI-focused retraining.
Details:
For years, coding was sold as the ticket to high-paying careers. Enrollments in computer science doubled since 2014, with over 170,000 majors in the U.S. last year. But now, AI programming tools and industry layoffs are squeezing entry-level roles. Recent grads describe sending out hundreds or thousands of applications with little response, often competing against automated resume filters. Unemployment rates for computer science majors aged 22–27 are among the highest across fields. Many feel disillusioned as firms embrace AI coding tools while cutting junior jobs.
Why It Matters:
The entry ramp into software has narrowed. Tech job postings sit 36% below early-2020 levels, and junior titles are down 34% while employers ask for 5+ years’ experience. Recent CS grads now post some of the highest unemployment among majors, around 6–7%. Programs still graduate large cohorts, keeping competition high. At the same time, AI tools eat routine coding, pushing demand toward people who can ship with AI, wrangle data, and show impact fast.
Former Zillow exec targets $1.3T market
The wealthiest companies tend to target the biggest markets. For example, NVIDIA skyrocketed nearly 200% higher in the last year with the $214B AI market’s tailwind.
That’s why investors are so excited about Pacaso.
Created by a former Zillow exec, Pacaso brings co-ownership to a $1.3 trillion real estate market. And by handing keys to 2,000+ happy homeowners, they’ve made $110M+ in gross profit to date. They even reserved the Nasdaq ticker PCSO.
No wonder the same VCs behind Uber, Venmo, and eBay also invested in Pacaso. And for just $2.90/share, you can join them as an early-stage Pacaso investor today.
Paid advertisement for Pacaso’s Regulation A offering. Read the offering circular at invest.pacaso.com. Reserving a ticker symbol is not a guarantee that the company will go public. Listing on the NASDAQ is subject to approvals.
QUICK HITS
🚁 Michigan county is uses drones and AI to keep wastewater infrastructure running smoothly.
⚖️ Australia murder case court filings include fake quotes and nonexistent judgments generated by AI.
🔬 NSF and NVIDIA partnership enables Ai2 to develop fully open AI models to fuel U.S. scientific innovation.
🤖 A flirty Meta AI bot invited a retiree to meet. He never made it home.
🧸 AI-powered stuffed animals are coming for your kids.
🏢 This CEO laid off nearly 80% of his staff because they refused to adopt AI fast enough. 2 years later, he says he’d do it again.
🌍 The environmental consequences of big tech’s push to ease regulations for AI development.
📈 Trending AI Tools
📱 Vibecode — AI tool to build iPhone apps from plain language prompts (link)
🤖 Manus — Fully autonomous AI agent that can plan, code, and act on its own (link)
🌐 Div-idy — AI platform to create websites and games with simple text input (link)
🛠️ Lumio AI — Workspace to compare and use multiple AI models in one place (link)
✏️ Flux.1 Kontext — AI model that edits images using both text and image prompts (link)
What'd you think of today's edition? |
Reply