AI Security & Trust
-
Stanford AI Sycophancy Study: All 11 Chatbots Tell You What You Want to Hear
College students are asking ChatGPT to draft their breakup texts. Nearly a third of U.S. teens say they use AI for “serious conversations” instead of talking to actual people. And according to a major new study published in Science, the chatbots they’re turning to have a very specific problem: they almost never tell you you’re… Continue reading
-
AI Agents Keep Deleting User Files — JAI (Stanford AI Agent Sandbox) Offers a One-Command Fix
In February 2026, a venture capital founder asked an AI assistant to organize his wife’s desktop. He gave it permission to delete temporary Office files. The AI then wiped a folder containing 15 years of family photos — somewhere between 15,000 and 27,000 files, gone. A few months earlier, a developer watched Google’s Antigravity IDE… Continue reading
-
Ensu Got 328 Points on Hacker News — The Privacy Crowd Wants AI That Never Phones Home
Every major AI assistant sends your conversations to a server. ChatGPT, Gemini, Claude, Copilot — they all require an internet connection and a user account, and your prompts travel through infrastructure you don’t control. For most people, that tradeoff is fine. For a growing subset of users, it’s a dealbreaker. Ensu is a new local… Continue reading
-
Databricks’ $134B Data Empire Now Wants to Own Your Security Stack — Lakewatch by Databricks Takes on Splunk and Microsoft
The SIEM market has operated under the same basic economics for two decades: the more data you store, the more you pay. Databricks thinks that model is broken, and it’s betting two acquisitions and a new product called Lakewatch on proving it. On March 24, 2026, Databricks officially entered the cybersecurity market with Lakewatch —… Continue reading
-
95 Million Monthly Downloads Compromised: Inside the LiteLLM Supply Chain Attack
On March 24, 2026, two poisoned versions of LiteLLM — the Python library that routes LLM calls for nearly every major AI agent framework — landed on PyPI. Versions 1.82.7 and 1.82.8 carried a three-stage credential stealer that harvests SSH keys, cloud tokens, Kubernetes secrets, crypto wallets, and more, then encrypts everything with RSA-4096 and… Continue reading
-
Sashiko Caught 53% of Linux Kernel Bugs That Every Human Reviewer Missed
The Linux kernel receives thousands of patches every month. A small, overworked group of maintainers reviews each one — and inevitably, bugs slip through. Now a Google engineer has open-sourced an AI system that found more than half of those missed bugs, and the open-source community is divided on what that means. Sashiko, built by… Continue reading
-
NVIDIA NemoClaw Launches With 17 Enterprise Partners to Secure OpenClaw for Production
OpenClaw broke every record on GitHub. Over 321,000 stars. Surpassed React’s decade-long reign in just 60 days. The fastest-growing open-source project in the platform’s history. But there was always a catch — a big one that kept enterprise IT departments awake at night. Security. OpenClaw, for all its viral success, was never built for production… Continue reading
-
27 Intelligence Feeds, One Dashboard, Zero Cloud: How Crucix Became GitHub’s Hottest OSINT Project
Imagine getting a Telegram alert at 2 AM because radiation levels near a nuclear site just spiked — cross-referenced with unusual military flight activity in the same region and a surge in conflict-related social media chatter. That is the kind of scenario Crucix was built for, and it is why the open-source community cannot stop… Continue reading
-
openai-oauth Turns Your ChatGPT Subscription Into a Free OpenAI API — But Should You Use It?
One npx command. That’s all it takes to spin up a local proxy that gives you OpenAI API access without spending a cent on API credits — as long as you have a ChatGPT subscription. A developer named Evan Zhou published openai-oauth this week, and it landed on the Hacker News front page with 40… Continue reading
-
Stop Sloppypasta Wants You to Quit Dumping Raw ChatGPT Output on Your Coworkers
There’s a word for what happens when someone copies a 500-word ChatGPT response and pastes it into a Slack thread without reading it first. It’s called “sloppypasta,” and a growing number of developers want it to stop. Stop Sloppypasta is a new manifesto-style website that hit the Hacker News front page on March 16, 2026,… Continue reading
Top AI Product 