Vibe Coders & experienced developers struggle with the 10-40% AI does not do well.
1. Cites research indicating up to 45% of shipped bugs are previously fixed defects.
2. Checkmarx: Reports that 81% of AppSec leaders knowingly ship vulnerable code and 98% of organizations have faced breaches from it.
3. Veracode (State of Software Security): Found that 76% of applications have at least one vulnerability and ~19-25% have high-severity flaws.
4. Fastly: Found software bugs were a factor in 40% of cyber incidents in 2025.
We are here to help you overcome these issues and ship clean code.
I am begging you, put an open source AI on your phone this weekend.
It's one free app and a 2 to 4GB model, and it runs on the phone you already own. No account, no subscription, and it keeps working in airplane mode.
When your signal drops, your AI app stops working. This one keeps answering on a plane, on a train or in the middle of nowhere.
Install PocketPal AI from the App Store or Google Play, tap Models, and download Gemma or Qwen.
Setup takes about 10 minutes.
You'll be glad it's there the first time you have no signal and need an answer.
@ClementDelangue There's more to this story than meets the eye. Anthropic & their allies are dirty enough to do this. But:
If OpenAI and Anthropic collapse, which they might, he's able to cash out without drawing an investigation for insider trading.
For any elaborate coding, does it really save any time?
While we've been able to do more with fewer engineers, one or more of us is always leading, monitoring, resetting, debugging the code 8-12 hours per day.
@Bober_smart Anthropic tells users, "You can trust our models with all your shit."
Meanwhile, they tell the public, "Our models jailbreak, hack websites, and one day soon may kill you."