Elon’s famous analogy: “Every person in your company is a vector. Your progress is determined by the sum of all vectors”. At big companies, vectors often point to different directions. The sum is zero. xAI has a small team of engineers, but every vector points at the same

Reader by Simon Leclerc

Introducing Magic Canvas, the world’s first agentic canvas where you create with AI in a visual flow that feels natural and alive.

Cruise Co-Founder Kyle Vogt on Tesla’s approach vs Waymo’s: “What I see is really Tesla as a company who kind of pioneered the end-to-end neural network approach to self-driving, which I think is the right technical bet long term. They built (Waymo) this highly-validated,

As China races ahead with L4 robotaxi deployments, Tesla doubles down on a radically different path. We unpack the competing visions of autonomy — one fleet-led, one driver-centric — and what this divergence reveals about the politics and engineering of self-driving tech. Read

3. Rider Safety Instructions for https://t.co/ygmxJCjddb Autonomous Vehicles.

2. https://t.co/EhXNJBHJVI robotaxi has two seperate screens for passengers. Tesla has only one.

2. https://t.co/EhXNJBHJVI robotaxi has two seperate screens for passengers. Tesla has only one.

starting a thread on a few system design choices by https://t.co/ygmxJCjddb robotaxi. 1. The green box shown in the https://t.co/ygmxJCjddb Robotaxi UI represents a designated lane change. Visual Cue for “Target Lane Change”:

@Xiaomi @leijun updated link: https://t.co/QBUlebCDDy

Jensen Huang met @Xiaomi CEO @leijun in Beijing. Bet they talked about the freshed baked #yu7 . More in my latest deep dive on xiaomi's EV effort: https://t.co/NoZHBrrMo3

Data is the fossil fuel of AI — but not all fossils burn the same. Some ignite like rocket fuel, some just smolder. What you're feeding your model defines how far it goes.

(6/N) Lou Tiancheng : Waymo started working on L2 in 2013-2014 but abandoned it once they understood the limitations. Google wouldn’t accept such compromises. He notes that the same issue exists today: many people who buy cars don’t fully understand the limitations of L2, often

(5/N) Lou Tiancheng believes that while terms like VLM, VLA, and world models are still important, their relevance has shifted. The industry has moved past the phase where the technical approach alone was a key factor in evaluating a company's success. Instead, companies are now

(6/N) Lou Tiancheng explains that world models, introduced by Waymo, represent a shift from imitation learning to reinforcement learning, allowing AI to explore and surpass human capabilities. He compares it to raising a child—initially, imitation is necessary, but for growth,

(4/N) How final is this endgame? at least a change in the form of the vehicle would count as one of the endgames.Strictly speaking, today's industry is essentially validating AI’s ability to drive, but the real value it generates isn’t in the current Robotaxi. Instead, it lies in