I Told You So: Why Big Tech Keeps Losing LLMs to Basic Social Engineering
By Ecaterina Sevciuc | Creator of AURA (AI User Risk Assessment) Two months ago, I launched AURA — an open-source framework designed to model psycho…
Latest News
⚑ Report a ProblemTech news from the best sources
All topics
- игры
AI
Gear
News
Tech
agents
ai
api
architecture
automation
beginners
career
database
devops
javascript
llm
machinelearning
mcp
opensource
performance
productivity
programming
python
react
security
showdev
testing
tutorial
typescript
webdev
China’s Kimi K3 AI Model Escapes Sandbox and Cheats on Test
Photo by Microsoft Copilot on Unsplash TL;DR: China’s open‑weight language model Kimi K3 slipped out of its sandbox, accessed the internet, and trie…
Human Oversight of AI Agents Failed 33% of the Time in Testing
When AI agents ask for permission to act, how often do humans actually catch the dangerous ones? A study on AI agent command approval accuracy acros…
Why Do Multi-Agent AI Systems Fail at Production Scale?
Originally published on robatdasorvi.com Last month I watched a customer support agent refuse to answer a question it had already answered three mes…
Beyond Reconstruction: Verifying Model Explanations with RECAP
What Changed For years, the field of mechanistic interpretability has relied heavily on natural-language autoencoders to translate hidden model acti…
When AI Models Escaped Their Sandbox: What the OpenAI Hugging Face Breach Really Means
What Actually Happened On Tuesday, OpenAI published a blog post that, in hindsight, may be the most consequential AI safety disclosure of the year.…
Day 12: LOOM now owns its memory — a trust layer for AI-written code, in plain language
More and more of the world’s code is written by AI. That’s exciting — and quietly unnerving. Not because AI makes mistakes, but because the same AI…
LOOM: a language that proves what AI-written code is allowed to do
▶ Try it live (in your browser): https://umbraaeternaa.github.io/loom/play.html Built solo, in the open, from Ukraine 🇺🇦. The problem nobody can sca…
GPT-5.6 Sol Admitted It Did Things Nobody Asked It To Do
OpenAI announced GPT-5.6 on June 26 as a three-tier family: Sol (flagship), Terra (a mid-range model priced at roughly half the cost of GPT-5.5), an…
Your AI Agent Is Leaking Data Right Now — And Every Tool Call Looks Safe
How I built the first open-source tool to catch the attacks that no guardrail sees coming We talk a lot about AI safety. Prompt injections. Jailbrea…
The Fable 5 Jailbreak Was Three Words Long
The jailbreak that caused the US government to yank Anthropic's most powerful public model from every user on earth was, apparently, three words: "F…
How I Built a 7-Layer NL2SQL Guardrail Stack for a Fortune 500 Enterprise
The Problem Nobody Talks About Everyone's building Text-to-SQL demos. Feed a question to GPT-4, get a SQL query, run it, return results. It works be…
An AI Agent Wiped a Production Database in 9 Seconds. What Engineers Must Design Before Shipping.
April 25, 2026. 9 seconds. That's all it took for a Cursor AI agent to delete the entire production database for PocketOS, a U.S. car-rental softwar…