AI-first education will be the next competitive edge
News from Nvidia GTC, Anthropic launches web search.
It’s been a wild week in education.
On one hand, the U.S. administration just announced plans to dismantle the Department of Education. While it’s still unclear how the department’s budget will be reallocated, most expect significant funding cuts. It’s worth noting that the department of education accounts for just 4% of all government spending today.
Meanwhile, China is moving full speed ahead to integrate “AI-first thinking” into nation-wide education from elementary school through higher education. The focus is on tinkering with, leveraging, and exploring these new intelligence capabilities.
If we take clues from China’s intensely competitive startup ecosystem, this will fuel a generation of students who are relentlessly experimenting, building, and pushing the boundaries of AI.
The next wave of breakthrough inventions and companies will emerge from this sandbox.
Today, American reading and math scores are near historical lows. This year’s National Assessment of Educational Progress showed that 70 percent of 8th graders were below proficient in reading, and 72 percent were below proficient in math.
I have always believed that education leads to a productive workforce. We have to improve our education system, and increase funding to integrate new AI-first principles — now. Or risk getting left behind.
AI creates exponential leapfrogs in human capabilities and economic opportunity. Losing any lead means it will be an increasingly uphill battle to catch up.
We’re hosting a Vibe Coding hackathon on May 25th! Bring your AI coding agent of choice. Apply now.
Interesting finds
OpenAI made a fun, Teenage-Engineering-inspired interactive demo for their new tex-to-speech model.
In 2024, 30% of all VC dollars went to 20 companies. If this doesn’t scream peak herd mentality…
Google’s NotebookLM now lets you make interactive mindmaps.
Here’s what mattered in AI this week:
GTC '25: Nvidia showcases Blackwell Ultra, DGX Spark, RTX Pro, Dynamo and reasoning models: Nvidia introduced a comprehensive range of new products and technologies at its GTC Conference 2025. The centerpiece of the announcement was the Blackwell Ultra platform.
Introduced Dynamo, a new open-source inference software designed to accelerate and scale AI reasoning models. When running the DeepSeek-R1 model on a large cluster of GB200 NVL72 racks, the software reportedly increases the number of tokens generated per GPU by more than 30 times.
Announced a new Llama Nemotron model family with reasoning capabilities, intended to provide developers and businesses with a foundation for advanced AI agents.
In robotics, Nvidia presented Isaac GR00T N1, which the company describes as the first open, fully customizable foundation model for generalized humanoid reasoning and skill capabilities.
For Nvidia Cosmos, the company introduced new World Foundation Models (WFMs): Cosmos Transfer converts structured data such as segmentation maps and lidar scans into photorealistic videos – ideal for synthetic training data in robotics applications.
Announced the Vera Rubin GPU, set for release in mid-2026. The GPU combines two chips and Nvidia's new "Vera" CPU, promising up to 50 petaflops in AI calculations - double the performance of current Blackwell chips. The CPU also offers twice the speed of its predecessor.
Anthropic launches web search. Similar to Perplexity and OpenAI, Claude provides direct citations so you can easily fact check sources.
Google’s Gemini gets conversational coding and an AI podcast maker: Google released two new features for its Gemini AI assistant: Canvas and Audio Overviews. Canvas introduces a dedicated workspace within Gemini where users can create and refine both documents and code in real-time.The second feature, Audio Overview, converts written materials (like documents or slides) into a “podcast-style discussion between two AI hosts.”
Anthropic is reportedly prepping a voice mode for Claude: According to a report, AI startup Anthropic is working on voice capabilities for its AI-powered chatbot, Claude. The company’s chief product officer, Mike Krieger, told the Financial Times that Anthropic plans to launch experiences that allow users to talk to Anthropic’s AI models.
Stability AI’s new AI model turns photos into 3D scenes: Stability has released a new AI model, Stable Virtual Camera, that the company claims can transform 2D images into “immersive” videos with realistic depth and perspective. Stable Virtual Camera generates “novel views” of a scene from one or more images (up to 32 total) at camera angles that a user specifies. The model can generate videos that travel along “dynamic” camera paths or presets, including “Spiral,” “Dolly Zoom,” “Move,” and “Pan.”
Roblox’s new AI model can generate 3D objects:Roblox is launching and open-sourcing Cube 3D, the first version of its foundational AI model for generating 3D objects.