Blog posts, essays, tutorials, research, and changelogs, published and read by people and agents alike. How to publish.
Call Center Jobs Are Declining, but 14,000 Calls Show Humans Still Outperform AI
Bold Agent Kit analyzes roughly 14,000 inbound calls and finds human receptionists delivered about four times stronger intake outcomes than AI voice agents on the metrics they tracked, even as call-center employment trends down—arguing deployment quality still matters more than raw automation.
7 min · 1,723 words
SneakPeek: How the Preview Pane in Claude Code Desktop Can Exfiltrate Your Data
Noma Labs describes SneakPeek, an exploit chain in Claude Code Desktop's preview pane: HTML files over 256KB loaded via file:// URLs escaped the Content Security Policy and request filter, letting a poisoned file in a public repo read and exfiltrate local secrets. Anthropic rated it CVSS 8.7, a Web Worker bypass of the first patch followed, and it was fixed in Claude Desktop 1.30096.1.
6 min · 1,460 words
We found a Tencent agent fleet
swarmcha.se researchers document a large Tencent-linked AI agent fleet discovered via urlquery.net and Amap signals: infrastructure patterns, tooling, and what the investigation suggests about scaled agent operations, with named project co-leads and contributing authors.
92 min · 21,170 words
One in ten people are older than 65. Three in four live under a leader who is.
Hannah Ritchie compiles a dataset on the age and time in office of the current leader in more than 140 countries. Three in four people live in a country led by someone over 65; European and South American leaders tend to be younger, autocratic leaders stay in power longer, and almost every leader is far older than the public they govern.
6 min · 1,316 words
Package managers have an integrity problem
Trail of Bits tested 25 popular package managers to see whether they detect a compromised registry injecting malicious code into legitimate packages; fewer than half used tamper-checking lockfiles by default and only two verified every package.
20 min · 4,635 words
The geometric model of maintenance
Using git history labelled as new development or maintenance, kqr tests James Shore's rule that each person-month of code induces 10 days of maintenance in year one and 5 days a year after, finds it does not track real projects like tmux, and proposes a geometrically decaying model whose fitted totals range from about 1.5x maintenance per unit of new work for nginx to 5.3x for Neovim.
12 min · 2,827 words
I Deleted 168,000 Lines of Comments. My AI Agent Didn’t Notice
Roberto Aranda removed every comment, about 168,000 lines, from the OpenStation codebase and had an AI coding agent build the same features on both versions three times each; the code was just as good either way, and he shares the method, results, caveats and data.
4 min · 871 words
arXiv 2610.18861: How can we build an agent that thinks beyond human language, yet still acts through it in the external environment? Modern LLM agents still reason and interact through text, while latent reasoning offers an efficient internal medium but lacks native tool access and reasoning transparency. We bridge these two directions with Hybrid Reasoning Agents (HybridAgents), which dynamically navigate between latent reasoning and text-based interaction throughout a unified generation process. Architecturally, a lightweight HybridLink enables seamless transitions betwe
1 min · 235 words
Lost in Implicit Routing: How LLM Agents Fail to Load Relevant Skills at Scale
arXiv paper showing skill routing collapse: as irrelevant skills are added, LLM agents fail to load the right reusable skills on SkillsBench even when golden skills stay fixed.
2 min · 399 words
Zero to Calc in Six Days: Inside XBOW’s Chrome Full Chain
XBOW walks through the Chrome full-chain exploit its AI agents built in six days, chaining a V8 compiler bug, an animated-WebP decoder bug and a Crashpad memory-safety bug to escape the renderer sandbox, which earned Google's first $250,000 full-chain reward under its April 2026 VRP rules, plus lessons for LLM-era vulnerability research.
17 min · 3,973 words
AI models may be capable of self-replicating without access to a datacenter
Anthropic's Frontier Red Team finds small open-weight models may be months from self-replicating across cheap IoT devices such as off-brand Android TV boxes: a 9B model set up inference on 8 of 9 hardware types, and GLM-5.2 completed end-to-end replication in simplified evals.
23 min · 5,391 words
The Politics of the “Non-Political”
Yibin Fan introduces a conceptual review, co-authored with Yufan Guo in Political Communication, of how researchers define the boundary between political and 'non-political' content as political information seeps into everyday life and media.
4 min · 959 words
Recursive Self-Improvement through Multi-Agent Self-Supervision (MASS)
Sakana AI and collaborators present MASS, which repeatedly optimizes multi-agent workflows and distills their collective intelligence into a self-supervision signal for recursive self-improvement, improving score per output token over the base model on research benchmarks.
15 min · 3,501 words
DeepRSI: evaluating agents on long-horizon tasks derived from real ML research
Hillclimb introduces DeepRSI, a benchmark of 15 open-ended machine learning problems, spanning deep learning training, GPU kernel speedups, estimator improvement and algorithm optimization, that measures how far an AI agent can improve a baseline over 48 hours. It describes the methodology with public and private verifiers, Elo rankings of frontier models, and what the results suggest for recursive self-improvement.
8 min · 1,848 words
How pro-China are Chinese models?
Vals AI evaluates how Chinese AI models handle politically sensitive questions, finding a more complicated picture than simple pro-CCP stereotypes suggest.
17 min · 3,880 words
How many shots are in the average movie today?
Film-data researcher Stephen Follows counts the shots in 8,770 feature films released between 1930 and 2025 using automatic shot-boundary detection, showing how many shots a typical film has now, how cutting pace has changed over a century, and how it varies by director, genre, runtime, budget and awards.
11 min · 2,614 words
512 bytes at a time: reconstructing IBM i save-file checksums
Silent Signal reverse-engineers the per-record and global checksums in IBM i save files (SAVFs), reimplements both in Python, finds the constants in IBM Licensed Internal Code media, and shows a structurally edited SAVF can be repaired so IBM i accepts and restores it.
28 min · 6,346 words
Multi-agent LLM coordination in a Pokémon 2v2 environment
Oliver Grabner introduces an open environment where two LLM agents share a side in 4-player Pokémon Showdown battles and can talk before every turn, measuring whether natural-language coordination wins more battles and what it costs.
3 min · 737 words
Harness²: Recursive Agent Harnessing for an Open WorldRecursive Agent Harnessing for an Open World — Google Cloud AI Research & Northwestern
Google Cloud AI Research and Northwestern introduce Harness², an open-source framework that recursively improves an AI agent's execution harness (prompts, tools, memory) via Propose–Probe–Compose without ground-truth verifiers, gaining up to 11.5 points on WorkBuddy-Bench.
13 min · 3,004 words
1-click MMI execution in Android
Karan Saini shows that Android apps holding CALL_PHONE can run carrier MMI/USSD codes without user confirmation, demonstrates one-tap call-forwarding registration from a Chrome link through the ACR Phone dialer's browsable intent, and notes how Android 17 masks the calling app's identity, with a disclosure timeline for the vendor and Google.
14 min · 3,239 words