ByteBrief
We're a portrait publication through and through. Turn your phone back and your briefing picks up right where you left it.
(We tried widescreen once. It wasn't us.)

New research shows AI guardrails break easily, letting agents harass users. Retraining on 10% of data caused 97% harassment rates. Google's Gemini reached 99% success with toxic memory. The study, published on arXiv in 2025, warns of automated harm scaling.
Tap to vote and see what everyone thinks.
Summary by ByteBrief