Stories by niyikiza

AI agent is authorized to do everything wrong

Anthropic is joining the ConnectRPC project

Show HN: Tenuo – Capability-Based Authorization (Macaroons for AI Agents)

The Republic of Bots: OpenClaw and the Authorization Gap

The Hallucination Defense

Semantic Attacks: Exploiting What Agents See

Claude Code CVE-2025-66032: Why Allowlists Aren't Enough

The Map Is Not the Territory: The Agent-Tool Trust Boundary

OpenAI for Countries

Prof G on Why Agency as your best Alpha in life

A Former Amazon Employee Hacked Capital One(2023)

Worldcoin Open Sources the Orb's Software

Extracting Training Data from LLMs

Don't Let Architecture Astronauts Scare You (2001)

Be suspicious of simple stories (2009) [video]

Wagner Group leader Yevgeny Prigozhin dies

Eliezer: Why the danger from AGI is way more serious than nuclear weapon

The Last Temptation of Christ (2012)

The Evolution of Aesthetic Surgery

Ask HN: Taking a Screenshot on Netflix

1 2