721
4TB of voice samples just stolen from 40k AI contractors at Mercor
Lapsus$ posted Mercor on its leak site. Voice biometrics plus ID documents in one bundle is a deepfake-ready kit. Here is what to do.
Top posts
Every blog post that took off on Hacker News, ranked by the votes of its best submission. 5,935 posts.
722
.gitignore Isn't the only way to ignore files in Git
You can ignore files in .gitignore, .git/info/exclude, and ~/.config/git/ignore
723
Americans' social activity has steeply declined in the past 30 years
Our survey finds significant, double-digit declines in Americans’ participation in key social activities from 1995 to 2026.
724
We need a federation of forges
Git is decentralized, but what of the rest?
725
A Jev that can ask Opus for help gives Opus accuracy at 4x lower cost
We tested Jev and GPT-6 Luna as judges for identity resolution. A hybrid of Jev and Claude Opus matched frontier accuracy at over 4x lower cost.
726
727
Incident CVE-2026-LGTM
A series of unfortunate agents.
728
Cursor launches Origin, GitHub alternative
Origin begins rolling out today in early beta on all paid plans. We're starting with the essentials, designed for agent scale: repos, pull requests, code browsing, and GitHub sync. Agent-native features ship soon.
729
Web Search API
Search the web from your AI agents and applications through AI Gateway with Ceramic.ai, Exa, and Linkup.
730
Qwen3.6-Plus: Towards real world agents
Qwen offers comprehensive functionality spanning chatbot, image and video understanding, image generation, document processing, web search integration, tool utilization, and artifacts.
731
Compromised LiteLLM 1.107.0 on PyPI Installs a Credential Stealer and Backdoor
LiteLLM 1.107.0 on PyPI ships a .pth credential stealer and a sitecustomize.py backdoor that run on every Python start. How the implant got in, what it steals, and indicators of compromise.
732
Plan mode is dead
Earlier this year, I believed planning was going to become the most important part of building software with AI.
733
If you're a button, you have one job
A blog about software craft and quality
734
Mistral X Mozilla: Private, Multilingual AI Browsing
Open, private and multilingual AI is coming to your web browser. Mistral and Mozilla team up to put powerful, trustworthy AI where you already browse.
735
How I find problems to solve as a staff engineer
How listening broadly turns everyday problems into useful project ideas.
736
Age verification for social media, the beginning of the end for a free internet?
So-called age verification for social media is spreading across the world, framed as an effort to create a safer internet for children. In reality, age verification lays the foundation for a fully government controlled internet.
737
Agents need control flow, not more prompts
Thesis: reliable agents tackling complex tasks need deterministic control flow encoded in software, not increasingly elaborate prompt chains If you’ve eve...
738
Introducing Ad Blocker for Firefox on iOS
There’s only so much room on your screen. Pop-ups, overlays, and ads can take over fast, getting between you and what you came to do. That’s where
739
Astra for Law
OpenAI for Law brings frontier intelligence for law, custom firm workflows, connected legal data sources, and legal-grade controls for confidential client work.
740
Record-High 89% in U.S. Say Government Corruption Widespread
Perceptions of government corruption in the U.S. hit a record high in 2026, with broad agreement across partisan lines.
741
Ember-1
Ember-1 is a new specialized model from Fireworks Research that delivers Kimi K3’s quality with 40% fewer tokens.
742
Bonsai 2 27B: Near-Lossless Compression in a 9x Smaller Footprint
Ternary Bonsai 2 27B retains 98.2% of Qwen3.8 27B benchmark performance in a 5.9GB footprint, with multimodal and agentic capabilities.
743
What Is a Harness?
An agent harness is a piece of software that provides an environment for an AI model to operate within.
744
Exploiting the most prominent AI agent benchmarks
Every week, a new AI model climbs to the top of a benchmark leaderboard. Companies cite these numbers in press releases. Investors use them to justify valuations. Engineers use them to pick which model to deploy. The implicit promise is simple: a higher score means a more capable system.
745
OpenAI’s accidental attack against Hugging Face is science fiction that happened
This story is wild. The short version: OpenAI were running a cybersecurity test against an unreleased model, with the model’s guardrail features turned off. Rather than solve the test, the …
746
Motorola Signature 27 with GrapheneOS support
Meet motorola signature 27, Motorola's most advanced flagship smartphone ever with next-gen AI, industry-leading tech, & a distnctive design.
747
The bottleneck was never the code
The other month I finally ran an experiment we had been postponing for over a year at .txt.
748
Banned book library in a wi-fi smart light bulb
OverviewA long while back I had an idea to hack a WiFi smart light bulb to do something more useful to me. Actually, I had a few different ideas of things to do with them. One of these ideas was to mo…
749
LLM Architecture Gallery
A gallery that collects architecture figures from The Big LLM Architecture Comparison and related articles, with fact sheets and links back to the original sections.
750
Google fixed more Chrome bugs in June than over the past two years, thanks to AI
Chrome uses Gemini AI to automate vulnerability discovery, triage, and patching, accelerating updates to match modern security risks.