>
Trump Evacuated As Shots Fired at White House Correspondents' Dinner, Shooter Dead
When the Cost of Truth Is High, We--and AI--Lie
Vanishing Minds: UFO & Nuclear Scientists Are Dying or Disappearing
The Great Iran Lie and the Persian Gulf Catastrophe
Researchers Turn Car Battery Acid and Plastic Waste into Clean Hydrogen and New Plastic
'Spin-flip' system pushes solar cell energy conversion efficiency past 100%
A Startup Has Been Quietly Pitching Cloned Human Bodies to Transfer Your Brain Into
DEYE 215kWh LiFePO4 + 125,000W Inverter + 200,000W MPPT = Run A Factory Offgrid!!
China's Unitree Unveils Robot With "Human-Like Physique" That Can Outrun Most People
This $200 Black Shaft Air Conditions Your Home For Free Forever -- Why Is It Banned in the U.S.?
Engineers have developed a material capable of self-repairing more than 1,000 times,...
They bypassed the eye entirely.
The Most Dangerous Race on Earth Isn't Nuclear - It's Quantum.

The high speed internet subscriber base could more than double to ~20–25 million by end of 2026 and then ~50 million in 2027. This will be about $25-30 billion of revenue by the end of 2026 when including launch revenue. The 2027 revenue could be about $40-50 billion.
Cursor built its own model but was compute-starved. They have tens of thousands of enterprise customers and 1 million developers as users. They were generating about $2.7 billion annual runrate revenue but were paying more to Anthropic and OpenAI. If they can switch customers over to their own LLM or XAI Grok Code then their cost of goods will drop to profitability. The whole AI coding space is surging and demand is more than supply. They were on track to get to at $6 billion revenue by end of 2026 but with more compute to train and compute to serve their models they should have far faster growth. $10-30 billion per year annual runrate is possible. The SpaceX-XAI buying Cursor deal will likely be a huge win.