Ron Diver
@rondiver.bsky.social
I build and operate businesses where technology meets physical production. Owner of a fourth-generation trade bindery in Seattle. I write code, build systems, and make real things.
This isn’t AI policy. It’s a permission slip. It avoids every hard decision: no liability, no copyright line, no real safety requirements. It clears the path to build, scale, and ship. The core move is preemption. Deploy now. Let the courts clean it up later. bit.ly/4lHEzSU
bit.ly
A Nature Medicine study found ChatGPT Health undertriaged 52% of true emergencies. But the deeper problem isn't the model — it's that no system exists to track what happens after any AI makes a clinical recommendation. The failure layer nobody is watching: www.lumenhealth.ai/insights/cha...
ChatGPT Health Missed Half the Emergencies. That's Not Even the Real Problem. | LumenHealth
A Nature Medicine study found ChatGPT Health undertriaged 52% of emergencies. But the deeper failure — the one nobody is measuring — is what happens after any AI system makes a clinical recommendation...
lumenhealth.ai
Everyone's chasing the AI model layer. The most durable assets in this revolution aren't digital. They're physical infrastructure, trust, integration, and coordination. The operators who already have those are sitting on more leverage than they realize.
