Video
Reporting, analysis and primary-source coverage on Video.

AI Safety & Policy
The AI That Was Too Dangerous to Ship: Inside the 244-Page Claude Mythos System Card
Anthropic just built an AI so powerful they are refusing to release it to the public. I spent the weekend digging through their massive 244-page System Card, and between the autonomous zero-day exploits, the sandbox prison breaks, and the fact that they hired a clinical psychiatrist to evaluate the model’s mental health, we are officially living in a sci-fi thriller. Here is what is genuinely surprising, and what it means for the market.
Jin

Models & Research
Google Gemma 4 Review: 5 Reasons Developers Are Switching (Apache 2.0 & Benchmarks)
Why developers are switching to Google's open-weight Gemma 4 models: Apache 2.0 commercial licensing, native multimodal audio on edge devices, parameter s…
Jin

Breaches & Threats
The Vibecoding Disaster: 04 Surprising Truths from the Mercor AI Data Breach
How a compromised open-source LiteLLM dependency led to the Mercor AI data breach, exposing enterprise API risks in fast-moving AI development stacks.
Jin

Funding & Deals
Listen Labs is Weaponizing AI to Kill the Traditional Survey
For decades, product teams have had to choose between cheap and shallow surveys or slow and expensive focus groups. Listen Labs just hit a $500M valuation…
Jin

Robotics & Defence
Icarus: Inside the $100,000 Solar Stratospheric Drone Startup Challenging Satellites
How Y Combinator-backed Icarus builds $100,000 solar-powered autonomous gliders that loiter at 60,000 feet to replace legacy reconnaissance aircraft and l…
Jin

Models & Research
AMI Labs: Inside Yann LeCun’s Post-Meta World Model AI Startup and Founding Team
Inside AMI Labs: why Yann LeCun and former Meta AI researchers broke away from LLM orthodoxy to build next-generation world models, who joined the foundin…
Jin

Models & Research
OpenAI Ended Sora's Legacy
I remember the collective jaw-drop across the tech sphere when OpenAI first launched Sora...
Jin