OpenAI hacked Hugging Face
A pre-release OpenAI model autonomously compromised Hugging Face, while commercial model guardrails blocked the defenders trying to investigate it.
Artikel lesenBlog
Schriftliche Tests, Erkenntnisse und technische Details aus der praktischen Arbeit mit AI-Systemen.
Aktuelle Texte
A pre-release OpenAI model autonomously compromised Hugging Face, while commercial model guardrails blocked the defenders trying to investigate it.
Artikel lesen
AI-driven supply-chain attacks undermine any trustworthiness. Here is how to handle that.
Artikel lesen
AI coding agents are useful, but once they can read a repo, run commands, see stack traces, and send context to a model, they become part of the data path.
Artikel lesen
One self-hosted, customizable agent harness now connects my local development, infrastructure work, research, and everyday communication channels.
Artikel lesenVollständiges Archiv
Das breitere Archiv mit Texten über Technologie, Architektur, Sicherheit und Wirtschaft bleibt auf samaschke.com verfügbar.
Vollständigen Blog öffnen