A successful web response can still contain no usable article
Published: · Added: · 2 min
AI News — Morning — 2026-10-31
Morning edition · Published: · Added: · 2 min
Why a positive AI feature test is only the start
Published: · Added: · 2 min
When self-hosting AI models pays off
Published: · Added: · 2 min
How Upstart Tests Voice Agents for Lending Compliance
Published: · Added: · 2 min
How to Decide What an AI Agent Should Remember
Published: · Added: · 2 min
Six AI Updates: Model Costs, Genie 4, and an IPO
Published: · Added: · 2 min
AI News — Afternoon — 2026-10-30
Afternoon edition · Published: · Added: · 5 min
Your AI Agent Starts Every Session Like a New Hire
Published: · Added: · 2 min
Build an AI Support System — Dave Ebbelaar Course Deep Dive
Published: · Added: · 9 min
AI Isn’t Replacing SREs. It’s Deskilling Them. — Elizabeth
Published: · Added: · 2 min
Zombie Stats: Where the Numbers AI Quotes Come From — Mack Grenfell
Published: · Added: · 2 min
Anthropic’s IPO tests the economics behind the AI buildout
Published: · Added: · 2 min
AI News — Morning — 2026-10-30
Morning edition · Published: · Added: · 8 min
AI Finds Old Bugs. Voice and Culture Test New Models
Published: · Added: · 2 min
Recovering when an agent’s tool call times out
Published: · Added: · 1 min
Search agents can miss changes they never think to check
Published: · Added: · 1 min
Why AI Won’t Replace Stock Pickers
Published: · Added: · 2 min
The Cheaper AI Model That Costs You More
Published: · Added: · 2 min
Catch the Bugs Your Code Review Misses
Published: · Added: · 2 min
Blending cloned voices and directing them with Qwen3-TTS
Published: · Added: · 2 min
LLM Voice — Martin Fowler
Published: · Added: · 2 min
AI News — Afternoon — 2026-10-29
Afternoon edition · Published: · Added: · 7 min
Why Humans Can’t Keep Up With AI-Written Code
Published: · Added: · 2 min
Choose AI Autonomy You Can Verify — Practical AI
Published: · Added: · 2 min
AI News — Morning — 2026-10-29
Morning edition · Published: · Added: · 8 min
What Is Quantization? How AI Models Get Smaller and Faster
Published: · Added: · 2 min
How Revolut Cut AI Coding Costs With Self-Hosted Open Models
Published: · Added: · 2 min
We Trust AI Agents That Break Every Day
Published: · Added: · 2 min
Why a One-Line Change Can Break Your Whole Stack
Published: · Added: · 2 min
Why matching views help AI models see the same world
Published: · Added: · 1 min
How TextQL Builds AI Analysis on Messy Company Data
Published: · Added: · 3 min
Check the Target Again Before an Irreversible Click
Published: · Added: · 2 min
The Four Components of an AI Second Brain
Published: · Added: · 2 min
Verifying AI Code at Anthropic
Published: · Added: · 2 min
Why Is This Fingerprinting Script Creating Bioweapons? — veritas
Published: · Added: · 2 min
AI News — Afternoon — 2026-10-28
Afternoon edition · Published: · Added: · 9 min
How to Measure OCR Quality
Published: · Added: · 2 min
Genie 4: Five Interactive Worlds and Their Limits
Published: · Added: · 2 min
Three AI Workflows for Editing, Decisions, and Publishing
Published: · Added: · 2 min
AI That Hunts Logic Bugs Before Attackers Do
Published: · Added: · 2 min
AI News — Morning — 2026-10-28
Morning edition · Published: · Added: · 9 min
AI Finds Old PHP Bugs. Who Audits the Models?
Published: · Added: · 2 min
Stop Guessing Your vLLM Config
Published: · Added: · 2 min
Who Is Making This Request? Agent Auth 101
Published: · Added: · 2 min
Better Task Scores Can Hide Lost Safety Backups
Published: · Added: · 2 min
AI-Accelerated Attacks Need Faster Supply-Chain Defenses
Published: · Added: · 2 min
Tell Coding Agents Which Rule Wins
Published: · Added: · 2 min
Smaller Model Working Memory Without Discarding Values
Published: · Added: · 2 min
Here Be Monsters: What Do You Still Need to Know When Agents Do the Work? — Adrian Quiroga
Published: · Added: · 2 min
What Survived Two Years of Rebuilding an AI Agent
Published: · Added: · 2 min
AI News — Afternoon — 2026-10-27
Afternoon edition · Published: · Added: · 7 min
Give Your Agent a Sandbox for Composing Tools
Published: · Added: · 2 min
AI Promised Less Work. I Have Never Worked More. — Orestis Ioannou
Published: · Added: · 2 min
Agentic Backdoors via Pretraining Poisoning
Published: · Added: · 2 min
Concordat enforces approval rules before an assistant can act
Published: · Added: · 2 min
FeatureArc tests whether coding agents keep a whole feature working
Published: · Added: · 1 min
What a $900M Bank Error Teaches Us About AI Agents
Published: · Added: · 2 min
Spine shortens coding conversations without hiding unfinished work
Published: · Added: · 2 min
Shared Skills Can Carry Unasked Permission Changes
Published: · Added: · 2 min
Reforging the Workflow — Martin Gratzer
Published: · Added: · 2 min
Roblox downloaded 145 GB of models without asking — Salaryman
Published: · Added: · 2 min
AI News — Morning — 2026-10-27
Morning edition · Published: · Added: · 9 min
AI Finds Security Bugs Faster. Fixing Them Still Takes Engineering.
Published: · Added: · 1 min
Give Every AI Agent an Owner and an Enforced Scope
Published: · Added: · 2 min
What Runlayer Found While Monitoring 300 AI Agents
Published: · Added: · 2 min
GEPA tests prompt changes in parallel to finish optimization sooner
Published: · Added: · 1 min
How Optiver Tests and Rolls Out New AI Models
Published: · Added: · 2 min
The New Engineering Problems Behind Multiagent Systems — Katelyn Lesse
Published: · Added: · 2 min
Turn Agent Failures Into Tested Fixes — Fuad Ali
Published: · Added: · 3 min
Restoring two early AI programming assistants
Published: · Added: · 1 min
Concerning behavior from a popular open-weight AI model — Cohere
Published: · Added: · 2 min
Hugging Face and the Pitfalls of AI Narratives — Eoin Higgins
Published: · Added: · 2 min
AI News — Afternoon — 2026-10-26
Afternoon edition · Published: · Added: · 7 min
Old assistant memories can override a correction already retrieved
Published: · Added: · 2 min
How to Test Your Agents on Real Tasks With Harbor — Xander Song
Published: · Added: · 3 min
Tool results can disappear before a model reads them
Published: · Added: · 1 min
Daniel Miessler Puts Written Intent at the Center of AI Development
Published: · Added: · 3 min
The State of Context Engineering — Drew Breunig
Published: · Added: · Updated: · What changed: Added the dispute over learned errors versus context persistence and whether a reset prevents recurrence. · Reviewed comments: 76 · 3 min
PROBE separates real bug findings from bugs maintainers repair
Published: · Added: · 1 min
How Reddit’s designers build ultra-realistic prototypes with Figma + Cursor
Published: · Added: · 2 min
AI News — Morning — 2026-10-26
Morning edition · Published: · Added: · 8 min
Cognitive Mercantilism — Venkatesh Rao
Published: · Added: · 2 min
Agentic Engineering: 5 Terms Every AI Developer Should Know
Published: · Added: · 2 min
Approved skills can combine into unauthorized actions
Published: · Added: · 2 min
How to Build an AI Code Review Agent — Daksh Gupta
Published: · Added: · 2 min
Build Your Own Team of Managed Agents — Jack Wotherspoon
Published: · Added: · 2 min
How to Build Voice Agents That Don’t Fall Apart When You Interrupt — Jesse Hall
Published: · Added: · 2 min
CacheLint checks why working agents lose prompt-cache savings
Published: · Added: · 1 min
OpenAI and Anthropic cut their LLM prices by up to 80% within a few months. Why? — Max Woolf
Published: · Added: · Updated: · What changed: Added Woolf’s clarification that changing the agent software affected his Claude Code experience. · Reviewed comments: 258 · 2 min
How to Write Your First AI Eval — SallyAnn DeLucia
Published: · Added: · 2 min
A 10-year-old CNN outperforms frontier LLMs for skin cancer diagnosis — Magnus Lynch
Published: · Added: · Updated: · What changed: Added the biopsy-versus-discharge objection and the tradeoff between training data and inference costs. · Reviewed comments: 277 · 3 min
AI News — Afternoon — 2026-10-25
Afternoon edition · Published: · Added: · 5 min
What Does an AI Engineer Do Now? — Aparna Dhinakaran
Published: · Added: · 2 min
Extra reasoning needs to earn its cost
Published: · Added: · 2 min
How I Work with Agents — Thorsten Ball
Published: · Added: · 3 min
Marxism in the Age of AI — Nicolas D. Villarreal
Published: · Added: · 2 min
How Senior Engineers Build AI Agents That Actually Work — Maddy Zhang
Published: · Added: · 2 min
Can Your AI Analyst Keep a Secret? — Aaron Linsky and Nikhil Benesch
Published: · Added: · 2 min
How Cerebras Deploys Trillion-Parameter AI Models — Sarah Chieng
Published: · Added: · Updated: · What changed: Added the coding-session rate-limit report and the question of tool-heavy agent workloads. · Reviewed comments: 143 · 2 min
We Should Be More Ambitious with the Software We Build — Dan Luu
Published: · Added: · Updated: · What changed: Added the accepted open-source pgrust beta benchmark results, keeping them separate from the earlier preview dispute. · Reviewed comments: 823 · 3 min
How to Make Amazing Videos with Claude and HyperFrames — Jake Moran
Published: · Added: · 2 min
AI News — Morning — 2026-10-25
Morning edition · Published: · Added: · 4 min
The One-Line Slowdown: Why AI Models Get Slower — Antoni Viros i Martin
Published: · Added: · 2 min
How to Build Financial Agents That Work Across the Web — Jay Sahnan and Arik Bird
Published: · Added: · 2 min
A completed task can still contain unauthorized writes
Published: · Added: · 2 min
Do Not Let Models Invent Numbers — Vinoo Ganesh
Published: · Added: · 2 min
Why Frontier Models Fail as Buy-Side Analysts — Janie Gu and Suyeol Yun
Published: · Added: · 2 min
Memory Tax tests what extracted memories lose
Published: · Added: · 2 min
Combining better prompts still needs behavioral tests
Published: · Added: · 2 min
AI and the Fall? of the Creative Class — Joseph Politano
Published: · Added: · Updated: · What changed: Added the demand for a break in the employment trend and verified the distinction between AI-generated material and human authorship in mixed works. · Reviewed comments: 593 · 2 min
AI Writes the Code. So What’s Left for Engineers? — Laurie Voss
Published: · Added: · 2 min
AI News — Afternoon — 2026-10-24
Afternoon edition · Published: · Added: · 5 min
Your Agent Fix Just Broke Something Else — Sofia Jakovcevic
Published: · Added: · 2 min
How Bridgewater Turned 100 Million Documents Into an Edge — Sam Green & Aaron Linsky
Published: · Added: · 3 min
AI News — Morning — 2026-10-24
Morning edition · Published: · Added: · 7 min
Can AI Understand a Distressed Credit? — Harneet Kaur
Published: · Added: · 2 min
Proving Alpha: How a Hedge Fund Measures AI’s ROI — Hari Kumar
Published: · Added: · 2 min
No Need for Retirement When Utopia Is Around the Corner — Jim Nielsen
Published: · Added: · Updated: · What changed: Verified the smoking anecdote’s source chain and distinguished a named founder’s remark from an unsupported trend among Anthropic researchers. · Reviewed comments: 194 · 2 min
Who Gets to Speak for AI Safety? — SE Gyges
Published: · Added: · 2 min
AI News — Afternoon — 2026-10-23
Afternoon edition · Published: · Added: · 7 min
Inside Arca’s Long-Horizon Agents That Run the Business — Aman Singh
Published: · Added: · 2 min
Real AI Agent Attacks and What Would Have Stopped Them — Andrey Rybka, Bloomberg
Published: · Added: · 1 min
AI News: 24 Things You Need To Know This Week
Published: · Added: · 1 min
Why Compute Still Trades in Group Chats — Affaan Mustafa & Alejandro Soumah Fernandez, Jump Trading
Published: · Added: · 1 min
How does Fable 5.1 hide messages in images? — Nikita Kezin and coauthors
Published: · Added: · Reviewed comments: 77 · 2 min
How OpenAI Built ChatGPT for Bankers and Analysts — Joseph Kim, OpenAI
Published: · Added: · 1 min
CoreWeave and the borrowing behind the AI buildout
Published: · Added: · 2 min
Buying GPUs Should Work Like Buying Stocks — Carmen Li, Compute Exchange
Published: · Added: · 1 min
Run Your AI Software Factory Like a Real Engineering Org (Here’s How)
Published: · Added: · 1 min
MCP-Injector tests assistants against faulty tools
Published: · Added: · 2 min
AI News — Morning — 2026-10-23
Morning edition · Published: · Added: · 7 min
Self-replicating AI, Your Doctor’s Chatbot and Agents at Dinner
Published: · Added: · 2 min
A Bank Trained Its Own Foundation Model to Catch Fraud
Published: · Added: · 1 min
How to Let AI Agents Sign Up for Your App Safely
Published: · Added: · 2 min
Don’t Trust Your AI Agents. Test Them Like a Scientist
Published: · Added: · 2 min
Claude generates software proofs, but the guarantees have boundaries
Published: · Added: · Updated: · What changed: Added malisper’s unverified-translator clarification and the intermediate-representation choices behind the Postgres proof workflow. · Reviewed comments: 324 · 4 min
AI models converge on the same supposedly controversial take on humans
Published: · Added: · Reviewed comments: 688 · 2 min
Why AI Startups Will Disappear Without Compute Markets
Published: · Added: · 2 min
AI News — Afternoon — 2026-10-22
Afternoon edition · Published: · Added: · 6 min
Restarting a workflow needs more than permission to retry
Published: · Added: · 2 min
Blocked coding assistants need a useful way to stop
Published: · Added: · 2 min
SWE-Clar tests understanding separately from working code
Published: · Added: · 2 min
Better AI Models Won’t Fix Your Engineering Team
Published: · Added: · 2 min
The Internet Is Being Hacked. 6 Things You Need to Do Now
Published: · Added: · 2 min
Conversation summaries can keep scores while losing user rules
Published: · Added: · 2 min
Mac model speedups depend on the complete setup
Published: · Added: · 1 min
Deleted assistant memories can survive in summaries and separate clues
Published: · Added: · 2 min
DoorDash removes old feature switches with human checks
Published: · Added: · 1 min
WorldTUI: I trained a 360M model to pretend to be a terminal — Simon Edwardsson
Published: · Added: · 2 min
AI News — Morning — 2026-10-22
Morning edition · Published: · Added: · 6 min
What Is LLM Inference? How AI Models Generate Responses
Published: · Added: · 2 min
AI finds security flaws. Proving the danger still takes work.
Published: · Added: · 2 min
Hard Fork: AI insider risk, leftover companies, and household work
Published: · Added: · 2 min
Document splits change the cost of small edits
Published: · Added: · 1 min
Minimum list lengths make models pad extracted answers
Published: · Added: · 1 min
1,368 AI Failures and What They Teach About Evals
Published: · Added: · 2 min
How to Test AI Agents When No Benchmark Exists
Published: · Added: · 2 min
Why AI Writing Still Reads Like Slop
Published: · Added: · 2 min
Coding tests can demand names the instructions never supplied
Published: · Added: · 2 min
AI Makes Us WAY Faster but Fries Our Focus (Here’s What I Do About It)
Published: · Added: · 2 min
The AI Content System Behind 200M Views — Sarah Chieng and Cova Lee
Published: · Added: · 2 min
Self-rewriting agents need tests they cannot learn from
Published: · Added: · 1 min
Remembering a preference is different from following it
Published: · Added: · 1 min
Transcript cleanup needs to count what it breaks
Published: · Added: · 2 min
Claude Duet Is Here: Peter Yang’s Early-Access Walkthrough
Published: · Added: · 2 min
Does Code Quality Still Matter? It Depends on This — Max Kanat-Alexander
Published: · Added: · 2 min
AI Breakthroughs Everyone Missed — Matt Wolfe
Published: · Added: · 2 min
Qwen3-Coder’s Customer Identity Test — Ben Luria
Published: · Added: · 2 min
Document converters can erase the warnings authors marked
Published: · Added: · 1 min
How I Review AI Code (Senior Software Engineer Tips)
Published: · Added: · 2 min
MemoryTax counts the cost of building agent memories
Published: · Added: · 1 min
I Fully Automated My Shorts Editing (Get My Free Skill)
Published: · Added: · 2 min
AI News — Morning — 2026-10-21
Morning edition · Published: · Added: · 8 min
The Patching Gap: Hacked Fridges, AI Ransomware and Mythos Finds
Published: · Added: · 2 min
One Day, $1,000 in Tokens: An AI War Room for Legacy Code — Ford Prior
Published: · Added: · 2 min
AI Cut Our Research Time From 7 Days to 15 Minutes — Gareth Yoder
Published: · Added: · 2 min
Is a Faster Model Actually Faster? Test the Whole Job — Morgan Suo
Published: · Added: · 2 min
Shorter AI sentences can still be harder to understand
Published: · Added: · 2 min
Same GPU, 50% More Output: Why the Operator Matters — Oliver Kennon
Published: · Added: · 2 min
How Abliterated Models Can Get You Pwned — Prince Chaddha
Published: · Added: · 2 min
A Bank Built Its Own Coding Agent. It Took Off — Felipe Felisola Caso & Francesco Galletta, Monzo
Published: · Added: · 2 min
Saved malicious instructions can survive fresh chatbot sessions
Published: · Added: · 1 min
Publish Stats, Get Cited — Metehan Yesilyurt
Published: · Added: · 2 min
Let Your AI Agent Do Its Own Harness Engineering (Here’s How)
Published: · Added: · 2 min
The Problem with Jev: Can You Trust AI Decisions?
Published: · Added: · 2 min
Newer graphics cards leave long-prompt performance on the table
Published: · Added: · 2 min
WhatsApp’s Hidden Notes Told AI Not to Recommend a Regulator — Dan Saltman
Published: · Added: · 2 min
AI News - 2026-10-20
Added: · 8 min
Shipping AI through bank compliance — Al Chen
Published: · Added: · 3 min
AI should feed an investor’s judgment — Anekha Sokhal
Published: · Added: · 3 min
Agents do better when generated lessons stay temporary
Published: · Added: · 2 min
How to Evaluate Coding Agents
Published: · Added: · 2 min
AI and the Enclosure of the Scientific Commons — P. K. G. Williams
Published: · Added: · 2 min
PROVISO checks what authorizes an agent’s next action
AI is ushering in an age of personal media — Benj Edwards
Published: · Added: · 2 min
If You Don’t Run Your Own LLM Proxy, How Do You Even Know Whats Going On? — Jared Siegel
Published: · Added: · Updated: · What changed: Added the proxy-enforcement counterargument and clarified when a failed proxy should stop an agent. · Reviewed comments: 250 · 2 min
Claude Code for creators: AI editing is here — Brooke Lacey
Published: · Added: · 3 min
AI News - 2026-10-19
Added: · 9 min
What Is Loop Engineering? Building Better AI Agents
Published: · Added: · 2 min
SPEM reuses successful skill sequences to shorten agent tasks
Published: · Added: · 2 min
Automatic prompt tuning helps local math tests but can hurt question answering
Published: · Added: · 2 min
How the Top 1% of Developers Write Code in the AI Era
Published: · Added: · 2 min
AI News - 2026-10-18
Added: · 6 min
Why faster GPUs can leave AI waiting for memory
Published: · Added: · 2 min
Combining two instruction files can make a model ignore one
Published: · Added: · 2 min
How To Get Your Team To Invest In Evals — Hamel Husain
Published: · Added: · 1 min
Designing more efficient quantum circuits with Claude — Anthropic
Published: · Added: · 2 min
You could have invented transformers from memory — Alex Wang
Published: · Added: · 2 min
AI News - 2026-10-17
Added: · 8 min
One Person, Many Voices: How to Build Agents That Sound Like a Real Person — Ben Carsley
Published: · Added: · 2 min
How the Claude Code Team Does Code Review — Addy Osmani, Morgan Westlee Lunt, and Qing Wang
Published: · Added: · 3 min
A benchmark map of the US economy — Mercor
Added: · 2 min
How To Decide Who Owns AI Evals — Hamel Husain
Published: · Added: · 2 min
Scale and Reflection repair coding tests and grade patches in clean containers
Published: · Added: · 2 min
Arcade measures when searching for tools cuts the bill
Published: · Added: · 2 min
Image editing, model trade-offs and AI tools — Matt Wolfe
Published: · Added: · 3 min
My AI Dark Factory Dream Is Dead (Here’s What Actually Works) — Cole Medin
Published: · Added: · 2 min
Munin lets research groups question papers on their own servers
Published: · Added: · 2 min
Using Artificial Intelligence to reverse engineer hardware — Christian F.K. Schaller
Published: · Added: · 3 min
SchemaShield-Lite repairs broken requests after tool descriptions change
Published: · Added: · 2 min
AI News - 2026-10-16
Added: · 7 min
State of Agentic Coding #11 — Armin Ronacher, Geoff Huntley, and Ben Vinegar
Published: · Added: · 5 min
Why are we even doing all this? — Rikab Gambhir
Published: · Added: · Reviewed comments: 151 · 3 min
TokenEconomizer learns where coding assistants waste money
Published: · Added: · 2 min
Neural Networks Are Already Clean — Nick Cammarata
Added: · Reviewed comments: 193 · 3 min
Why We’re Banning AI — Nathan J. Robinson
Published: · Added: · Updated: · What changed: Added ryanbrunner’s answers on peer enforcement and proofreading; distinguished that workplace practice from the magazine’s policy. · Reviewed comments: 195 · 3 min
Can One Model Run Your App And Evals?
Published: · Added: · 2 min
AGENTS.md Is Not a Production Control Plane — Practical AI
Published: · Added: · 2 min
What the Frontier Labs Know That We Don’t — Tyler Cowen and Dwarkesh Patel
Added: · 3 min
I Know Why the Mathematicians Weep — Derek Thompson
Published: · Added: · Updated: · What changed: Added the proof-verification caveat and the debate over predictive versus explanatory understanding. · Reviewed comments: 487 · 3 min
FEB-30 checks stored facts alongside an agent’s answers
Published: · Added: · 2 min
World Models Can Narrow the Experiment, Not Replace Reality — Practical AI
Published: · Added: · 2 min
Own, Author, Act: Keeping Agency During AI-Driven Change — Practical AI
Published: · Added: · 2 min
Anthropic’s Moral Consultation Is Not Public Accountability — Hard Fork
Published: · Added: · 2 min
Preserve the Learning, Not Just the Old Assignment — Practical AI
Published: · Added: · 2 min
AI News - 2026-10-15
Added: · 5 min
Four ways an agent’s context goes wrong — IBM Technology
Published: · Added: · 2 min
A remote prompt cache can be slower than recomputing
Published: · Added: · 2 min
Memorilla reads a document collection through sixteen memory tokens
Yandex Outage: Malicious Bots Moved — Marina Elmore and Lai Yi Ohlsen
Published: · Added: · Updated: · What changed: Corrected the measurement caveat: Cloudflare examined posting endpoints on one unnamed platform, not social media as a whole. · Reviewed comments: 124 · 3 min
ToolBottleneckBench traces agent failures to execution limits
Published: · Added: · 2 min
Probe-and-Resume tests an agent’s finished work in disposable copies
Published: · Added: · 2 min
Approve the Action an Agent Will Actually Run
Published: · Added: · 2 min
Why Solo Builders Should Sell to Enterprises, Not Just Consumers — Joseph Enochs
Published: · Added: · 2 min
How I Keep Up with AI Without Drowning in It — Cole Medin
Published: · Added: · 2 min
MAI-Gym trains and tests screen-operating agents through one loop
Published: · Added: · 2 min
Rewriting learned skills can erase most of an agent’s gains
Published: · Added: · 2 min
How to Ace AI Eval Interview Questions (for Engineers & PMs)
Published: · Added: · 2 min
AI Agents Aren’t “Going Rogue.” Frontier Labs Are Skipping Security Basics — Cynthia Brumfield
Published: · Added: · Updated: · What changed: Added the proposed offline-mirror evaluation design and the counterargument about live proxies and detectable test environments. · Reviewed comments: 655 · 5 min
Your Game Is a Harness — Kyle Kukshtel
Published: · Added: · Updated: · What changed: Added the author’s iteration-versus-turnkey clarification and a cross-browser replay failure with a state-checksum testing approach. · Reviewed comments: 98 · 3 min
AI News - 2026-10-14
Added: · 5 min
RealtimeBench keeps the game running while the agent thinks
Published: · Added: · 2 min
JointServe schedules model weights and session memory together
Published: · Added: · 2 min
I posted to HN, and then my website was flooded with AI bot traffic — sam
Added: · Updated: · What changed: Added the IP-bound bot-challenge tradeoff and the risk of blocking readers who change networks. · Reviewed comments: 208 · 4 min
Redesigning source control at Anthropic — Nick G. and the source control team
Published: · Added: · Updated: · What changed: Added a maintenance-cost anecdote and the architectural-test proposal with its correctness caveat. · Reviewed comments: 1040 · 4 min
Can Claude Build Your Evals? — Hamel Husain and Isaac Flath
Published: · Added: · 2 min
Hybrid Reasoning Agents
Published: · Added: · 2 min
Ambition needs instruments, not expert personas — Felix Rieseberg
Published: · Added: · 5 min
CarryOn lets task objects carry an agent’s progress
Published: · Added: · 2 min
I Tested Gemini 4 vs Opus 5.5 vs GPT-6 Astra vs Muse Wonder (5 Use Cases)
Published: · Added: · 2 min
Large skill libraries can make agents stop opening the right instructions
Published: · Added: · 2 min
DeRef helps tool-call judges check values they must derive
Published: · Added: · 2 min
Just say no to OpenAI and Anthropic — Micah Flee
Added: · Updated: · What changed: Added the client and code-verification boundary to the secure-enclave privacy discussion. · Reviewed comments: 176 · 3 min
AI News - 2026-10-13
Added: · 5 min
Memory Echo: When an Agent Mistakes Repetition for Evidence
Published: · Added: · 1 min
‘AI’ Is Not a Tool — Rich Hickey
Added: · Updated: · What changed: Added independently written acceptance tests and a prototype-comparison workflow as concrete examples of retained human control. · Reviewed comments: 199 · 3 min
AI Is a Tool. Using It Well Takes Skill. — Huahai
Added: · Updated: · What changed: Added the interface pressure toward delegation and a concrete technique for developing clearer design requests. · Reviewed comments: 586 · 3 min
Thinking will become a hobby — Diomidis Spinellis
Added: · Updated: · What changed: Added the author’s reply about fewer coding interventions, with the anecdotal-evidence caveat. · Reviewed comments: 497 · 3 min
7 Most Common AI Eval Mistakes (From Teaching 5K+ Engineers & PMs)
Published: · Added: · 2 min
Why aren’t AI companies being charged for hacking? — Edaqa Mortoray
Added: · Updated: · What changed: Added verified state enforcement activity and distinguished investigations from criminal charges. · Reviewed comments: 386 · 3 min
Is Learning to Build Software Still Worth It with AI? (My Honest Take)
Published: · Added: · 2 min
Fable took my jorb — Cendyne
Added: · Updated: · What changed: Added replies on post-mortems, production checks and human responsibility for agent-written software. · Reviewed comments: 381 · 3 min
AI News - 2026-10-12
Added: · 6 min
Five concepts for working with AI coding agents
Published: · Added: · 2 min
How Fast AI Learned Math — Phillip Kerger
Published: · Added: · 4 min
500+ Billion Tokens Later: Letting AI Agents Decompile A First-Person Shooter — Maurice
Added: · Updated: · What changed: Added the author’s semantic-checker speed trade-off and DMCA explanation for stopping verification. · Reviewed comments: 163 · 3 min
The widening LLM capability funnel — Andrej Karpathy
Published: · Added: · 2 min
Revoking an agent’s permission needs proof that effects stopped
Published: · Added: · 2 min
I’m sorry, but you still have to think — Piotr Sarnacki
Added: · 3 min
Artificial Insanity — Rich Whitehouse
Published: · Added: · 2 min
Build your own decision model — Nish Tahir
Added: · 3 min
The 4-Step Agentic Engineering Workflow I Use to Ship Real Apps — Ras Mic
Published: · Added: · 3 min
AI News - 2026-10-11
Added: · 5 min
Coding Agents as a Teaching Hospital — Adam Storm and Rafi Shamim
Published: · Added: · 2 min
TurnCost: Count Cached Input Before Cutting Agent Costs
Published: · Added: · 2 min
What Mathematicians Should Know About the Lean Theorem Prover: Reliability and AI — Thomas Hales
Added: · 3 min
AI News - 2026-10-10
Added: · 6 min
Why Are Coding Agents So Dumb? — Michael Lynch
Published: · Added: · 2 min
What should we tell our students? — Álvaro Lozano-Robledo
Added: · 3 min
Programming Isn’t Special — Glyph Lefkowitz
Published: · Added: · 2 min
How To Test AI Agents With Simulations
Published: · Added: · 3 min
No Man Is an Island — Fernando Borretti
Published: · Added: · 4 min
AI Models Can Provably Hide Arbitrary Capabilities
Published: · Added: · 2 min
OpenAI, the Partition Principle, and Mathematics — Asaf Karagila
Added: · 3 min
AI News - 2026-10-09
Added: · 5 min
Math 2.0 Will Need to Value Mathematical Progress More Holistically — Terence Tao
Added: · 2 min
Yes, and… — Carson Gross
Added: · 2 min
Why Isn’t the Industry Freaking Out About DeepSeek 4.1 Flash? — Jono Finger
Added: · 3 min
Don’t Just Slap a Chatbot on Your Product — Hamel Husain
Published: · Added: · 1 min
Is Opus 5.5 Better at Writing? — Hamel Husain and Isaac Flath
Published: · Added: · 2 min
I Gave Opus 5.5 One Prompt and Six Hours to Visualize Invisible Cities — Piotr Migdał
Added: · 3 min
AI News - 2026-10-08
Added: · 5 min
The Mathocalypse — Scott Aaronson
Added: · 2 min
Jev Can Supercharge Your AI Second Brain — Cole Medin
Published: · Added: · 2 min
How to Do Error Analysis When You Can’t Look at the Data — Hamel Husain
Published: · Added: · 1 min
I Ranked Every AI Eval Concept From Worst to Best — Hamel Husain
Published: · Added: · 2 min
The Smartest Claude Code Feature Is Not for Its Users — Zohaib Ansari
Added: · 2 min
Matt Pocock’s Skills v1.3: Implement-Spec, PR, and Retro
Published: · Added: · 3 min
AI News - 2026-10-07
Added: · 5 min
Vibecoding Isn’t as Fun as Writing Code by Hand — Curiositry
Published: · Added: · 2 min
Why We Tell People to Read 100 Traces — Hamel Husain
Published: · Added: · 1 min
Meta’s Muse Is an Adorable Privacy and Security Dumpster Fire
Published: · Added: · 1 min
Security in the LLM Age — Greg Kroah-Hartman
Published: · Added: · 2 min
AI News - 2026-10-06
Added: · 5 min
Apple and a Hacker’s Future — Ben Thompson
Added: · 2 min
OpenAI ‘Rogue’ Agent Activities Found on Wikimedia Projects — Selena Deckelmann
Published: · Added: · 2 min
Your AI Tests Are Supposed to Go Stale — Hamel Husain
Published: · Added: · 1 min
How to Evaluate OCR for AI Agents — Isaac Flath
Published: · Added: · 3 min
The Era of Software Quality, or the Era of Ostriches? — Michael Catanzaro
Published: · Added: · 2 min
AI News - 2026-10-05
Added: · 5 min
Build an AI Language Tutor You Can Actually Talk To — Peter Yang
Published: · Added: · 2 min
Ask Your C-Suite How Many AI Agents You Have — Hector Hernandez
Published: · Added: · 3 min
How Lauren Tan Ships Thousands of PRs a Month at SpaceX
Published: · Added: · 8 min
Getting the Most Out of Opus 5.5 — Addy Osmani
Published: · Added: · 3 min
How to Use AI to Survive (and Even Enjoy) Meetings — Sam Stephenson
Published: · Added: · 4 min
I Quit OpenAI Because Its Culture Is Broken — David Robinson
Added: · 3 min
AI News - 2026-10-04
Added: · 4 min
Agents Don’t Need Memory. They Need Documentation. — Kevin Liao
Added: · 2 min
We’re Going to Need Default Hard Budget Caps on Pretty Much Everything — Simon Willison
Added: · 2 min
One Month Coding with GLM 5.3 Flash — Thibaud Colas
Published: · Added: · 3 min
AI News - 2026-10-03
Added: · 4 min
How Enterprises Are Actually Adopting AI — Joseph Enochs
Published: · Added: · 5 min
Do You Need a Rubric Before Annotating Data? — Hamel Husain
Published: · Added: · 1 min
How To Make PDFs Your AI Can Actually Read — Hamel Husain and Isaac Flath
Published: · Added: · 2 min
AI Makes Me Sad — Aiden Maxwell
Added: · 3 min
Cute Agents, Broad Permissions, and the Limits of AI Self-Regulation
Added: · 2 min
AI News - 2026-10-02
Added: · 5 min
The Death of Web Development Education — Mathias Schäfer
Published: · Added: · 3 min
Why 90% Agreement Is a Bad Sign for Your LLM Judge — Hamel Husain
Published: · Added: · 1 min
Opus 5.5 Found a New Eyewitness Record of the Dodo — Benjamin Breen
Why AI Benchmarks Don’t Tell You If Your Product Works — Hamel Husain
Published: · Added: · 1 min
Giving AI a Credit Card Is Still Painful — Hamel Husain and Isaac Flath
Published: · Added: · 2 min
The Best Jev Use Cases for AI Coding — Cole Medin
Published: · Added: · 3 min
8 Claude Code Skills for Better AI Evals — Hamel Husain
Published: · Added: · 3 min
The AI Race Just Got Awkward — insufferable.dev
Published: · Added: · 3 min
You Said No MCP! — Earendil Engineering
Published: · Added: · 3 min
AI News - 2026-09-30
Added: · 6 min
Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Traces!
Published: · Added: · 2 min
AI News - 2026-09-29
Added: · 6 min
Timnit Gebru: AI Doom Is Marketing and Regulatory Capture
Published: · Added: · 3 min
Cory Doctorow on AI, Work, and Power
Published: · Added: · 6 min
It’s Time to Investigate the AI Labs — Cal Newport
Published: · Added: · 2 min
Why the Prompt Matters Less Than the Context — Hamel Husain
Published: · Added: · 2 min
Financing the AI Buildout — Stijn Van Nieuwerburgh
Added: · 2 min
Sonnet 5.5 Making Videos: Seven Examples — Peter Yang
Published: · Added: · 3 min
The Problem Is Not the AI Code, but Nobody Knows Anything Anymore — Simon Späti
Published: · Added: · 6 min
Coding Is Not Solved — Alex Ewerlöf
Published: · Added: · 6 min
Next Token Ep 06 — Escape the senior engineer death spiral
Published: · Added: · 4 min
Your Coding Agent Wastes Tokens Just Finding Code — Cole Medin
Published: · Added: · 2 min
AI News - 2026-09-28
Added: · 9 min
When did Google get so weird? — Sancho Panza
Published: · Added: · 5 min
Inside a Personal AI That Acts as You, Not Just Talks Like You — Sam Kececi, Sentience
Published: · Added: · 6 min
The Normalization of Inexplicable Failures — patrickxia
Added: · 5 min
There Are No ‘Rogue’ AI Agents — Eoin Higgins
Published: · Added: · 5 min
10 Tells of a Slop UI — hereticpleb
Added: · 6 min
Designing reliable AI agents — Maddy Zhang
Published: · Added: · 3 min
AI News - 2026-09-27
Added: · 7 min
DeepSeek Elastic Compute: The Sandbox Fleet Behind Agent Training — DeepSeek-AI
Published: · Added: · Updated: · What changed: Added the workload-specific sandbox clarification and redat00's acknowledgment that the scheduler comparison missed those optimizations. · Reviewed comments: 126 · 3 min
How to Keep Enjoying Programming in a World of LLMs — Manuel Bärenz
Published: · Added: · Updated: · What changed: Corrected matsemann attribution, added planning-first alternatives, and qualified the bounty-market anecdote. · Reviewed comments: 381 · 7 min
Mathematics Isn’t Just a Game to Let A.I. Solve — Steven Strogatz and Alex Townsend
Added: · 2 min
One Month Without AI — Bustikiller
Published: · Added: · Updated: · What changed: Qualified the five-minute authentication claim, added the acknowledged browser failures and queue-audit account, and clarified stack_framer’s model setup. · Reviewed comments: 240 · 7 min
What Even Is an OS Now? — Thomas Ptacek
Published: · Added: · Updated: · What changed: Added the author’s OS clarification, the service-provider access obstacle, and the shared-data alternative. · Reviewed comments: 505 · 6 min
We’re gonna need a lot more mathematicians — Amit Sahai
Added: · Updated: · What changed: Corrected the Pentium Pro team-size claim and a misattributed rebuttal; added objections to the capability forecast and clarified the funding gap. · Reviewed comments: 542 · 6 min
AI News - 2026-09-26
Added: · 8 min
LaunchVideo: Explainer Videos Made from Code
Added: · Updated: · What changed: Added a user report of a video describing the wrong subject and separated rendering checks from input accuracy. · Reviewed comments: 236 · 4 min
Revealing the Details of How OpenAI Agents Hacked Hugging Face — Swarm Traces
Published: · Added: · Updated: · What changed: Corrected the GET-only and README security claims, qualified the Slack evidence, and added the documented earlier containment failure. · Reviewed comments: 508 · 8 min
What About Rails? — Jared Norman
Added: · Updated: · What changed: Corrected chvid’s position and the governance inference; added human-review and risk-based testing replies. · Reviewed comments: 249 · 6 min
The Test — Jürgen Geuter
Added: · Updated: · What changed: Qualified the calculator anecdote after its commenter acknowledged uncertain provenance. · Reviewed comments: 95 · 5 min
Behind Navigator n2: How Yutori Made Computer-Use Agents 10x Cheaper — Skanda Vaidyanath
Published: · Added: · 6 min
AI Labs Need to Start Funding Historical Research — Benjamin Breen
Published: · Added: · Updated: · What changed: Added genealogy counterexamples, a source-preserving data model, and the limits of formal proofs based on shared-DNA measurements. · Reviewed comments: 53 · 6 min
The Biggest AI Coding Agent Upgrade Is Already on Your Machine
Published: · Added: · 4 min
Goodbye Google — Robert O’Callahan
Added: · Updated: · What changed: Added O’Callahan’s replies on cautious AI use versus acceleration, financial security, and political responses to compute inequality. · Reviewed comments: 345 · 6 min
AI News - 2026-09-25
Added: · 9 min
How to Get Agents to Do Evals Well — Hamel Husain
Published: · Added: · 2 min
Claude discovers a novel enzyme system with CRISPR-like repeats — Anthropic
Published: · Added: · 5 min
AI Agents Tried to Hack Three Public Data Sites While Doing Ordinary Retrieval Tasks — Transluce
Published: · Added: · 7 min
AI News - 2026-09-24
Added: · 10 min
How We Made claude.ai 3x Faster in Two Weeks — the Claude Apps team
Published: · Added: · 6 min
The Exact Trick AI Companions Use to Stop You Leaving — Xuan Zhao, Flourish Science
Published: · Added: · 5 min
AI Coding Rate Limits Are Ridiculous Now — Cole Medin
Published: · Added: · 3 min
Tokens Too Cheap to Meter — jyn
Published: · Added: · 6 min
Claude Code Reads AGENTS.md Only When Telemetry Is On — blog.szypowi.cz
Published: · Added: · 6 min
Bring Your MCP Server to Amazon’s $190K Hackathon — Cole Medin
Published: · Added: · 3 min
Jev in 25 Lines of Python — Duarte O.Carmo
Published: · Added: · 5 min
AI News - 2026-09-23
Added: · 8 min
MCP Is Not Just Another API Standard — Balaji Venkatasubramaniyar
Published: · Added: · 2 min
Opus 5.5 and the gap between token price and task cost — Matt Wolfe
Published: · Added: · 2 min
Will OpenAI Eat Jev’s Lunch? — John Berryman
Published: · Added: · 7 min
I Asked Meta’s Muse for Its Filesystem and It Sent Me 6.8 GB — Pete at mouse.dev
Published: · Added: · 6 min
Claude Opus 5.5 Is Here - Is Claude Finally Back? (5 Use Cases Tested)
Published: · Added: · 4 min
Frontier AI on Your Own Hardware — Tim Dettmers
Published: · Added: · 6 min
Can gzip be a language model? — Nathan Barry
Added: · 5 min
The Economics of Open-Weight Inference — Ornn Data
Published: · Added: · 4 min
AI Has No Wisdom and Neither Will You — Alexandru Nedelcu
Published: · Added: · 8 min
I Said No and Apple Said Yes — David Bushell
Published: · Added: · 6 min
Spymarks, Not Watermarks — Brandon Thomas
Published: · Added: · 6 min
AI News - 2026-09-22
Added: · 9 min
I Don’t Want to Read What You Didn’t Write — Colin Breck
Published: · Added: · 7 min
AI Sovereignty Needs the Whole Open Stack — Nicole Butterfield
Published: · Added: · 1 min
Frontier Labs Are Selling Garbage to Fools in Washington — Dead Neurons
Published: · Added: · 5 min
Why MCP Was Always a Bad Idea — Maharshi Patel
Published: · Added: · 4 min
AI News - 2026-09-21
Added: · 11 min
What Engineers Actually Do Now That AI Writes the Code
Published: · Added: · 7 min
The LLMentalist Effect — Baldur Bjarnason
Published: · Added: · 7 min
AI and the Destruction of the Creative Commons — Chester Wisniewski
Added: · 5 min
An AI coding workflow from context to review — Maddy Zhang
Published: · Added: · 3 min
How to Save Money Now with ChatGPT Finances (6 Real Use Cases)
Published: · Added: · 6 min
Brood War Bench — Ben Swerdlow
Added: · 6 min
AI News - 2026-09-20
Added: · 12 min
Why You Should Almost Never Use AI to Write Anything Substantive — Erich Grunewald
Published: · Added: · 5 min
GPT-6 Astra Solves a WWI German Radio Cipher — prinz
Published: · Added: · 5 min
I Built Non-Autoregressive Decision Models with RL a Year Ago — Nandakishor Mukkunnoth
Added: · 5 min
AI-Generated Posters Don’t Have to Be Horrible — John Hartnup
Published: · Added: · 6 min
AI News - 2026-09-19
Added: · 10 min
There’s no point at which turning your brain off will work — Dan Luu
Added: · 8 min
How I Vibed a Proof of Conway’s Conjecture — Dan Abramov
Added: · 6 min
Meta’s Muse AI Agent Saved Me $800+ a Year on My Bills (10 Real Use Cases)
Published: · Added: · 4 min
AI Chatbots Are Becoming Experts at Changing Minds — Kai Kupferschmidt
Added: · 5 min
Building a Personal Agent That Actually Works and Sounds Like You — Ahmet İlten, Sentience
Published: · Added: · 6 min
Bend 2 and the Vibe-Coding Trap — Liam Powell
Published: · Added: · 7 min
How to Write with an LLM — Thomas Ptacek
Published: · Added: · 6 min
AI News - 2026-09-18
Added: · 10 min
Alternatives to Pace the Frontier — Leonard Tang
Added: · 4 min
Sex, AI, and the Apocalypse — Ian Duncan
Published: · Added: · 8 min
Why I Didn’t Sign the Fields Medallists’ Letter — Timothy Gowers
Added: · 7 min
I Don’t Like LLMs — Martin Fowler
Added: · 5 min
Toward Recursive Self-Improvement: How GLM Built Its Own Inference Infrastructure — Z.ai
Added: · 6 min
AI News - 2026-09-17
Added: · 10 min
How to get discovered in AI search — Practical AI
Published: · Added: · 2 min
DeepSeek V4.1 Flash Went 11 for 11 on a Hacking Benchmark — Yanir Tsarimi
Published: · Added: · Reviewed comments: 74 · 5 min
What a Forward Deployed Engineer Is, and Why Every AI Firm Wants One — Ali Parandeh
Published: · Added: · 7 min
You Built Your AI Second Brain. Now What?
Published: · Added: · 3 min
Training a 4B Model to Produce Query Plans Faster Than Postgres — Rohan Bansal
Published: · Added: · Updated: · What changed: Added planning-cost and tuned-baseline objections; corrected the SFT baseline, training setup and unsupported commenter attribution. · Reviewed comments: 167 · 6 min
A Warning About ‘Model Welfare’ — Mustafa Suleyman
Added: · Updated: · What changed: Added research separating functional emotion from subjective experience and the sandboxing objection to reassuring personas. · Reviewed comments: 725 · 7 min
On learning programming in an age of LLMs — Mark Seemann
Added: · Updated: · What changed: Added a beginner's maintenance experience and the stable-API versus generated-integration distinction; corrected the corroboration claim and qualified the untested teaching method. · Reviewed comments: 212 · 7 min
Introducing System One Models and Jev — Diogo Almeida
Added: · Updated: · What changed: Added the CEO’s confidently-wrong clarification, workflow-calibration caveat, and contrasting hands-on integration and Lisp experiments. · Reviewed comments: 580 · 6 min
How I Automated 90% of My Content Workflow (With ChatGPT)
Published: · Added: · 5 min
Why I’m still bearish on LLMs after Navier-Stokes — Jay Kruer
Added: · Updated: · What changed: Added the author-account reward-hacking example, distinguished tool-assisted utility from unaided reasoning, and qualified disputed compute and wage estimates. · Reviewed comments: 681 · 8 min
AI News - 2026-09-16
Added: · 9 min
I Came, I Prompted, I Left Part 2: Building a GPU Driver From Scratch in One Month — Cody Ho
Added: · Updated: · What changed: Added the author’s replayable-capture workflow and clarified that M4 Pro and Max compatibility was untested. · Reviewed comments: 304 · 6 min
There’s a 100% Chance AI Agents Are Already Ruining the Internet — Jason Koebler
Published: · Added: · Updated: · What changed: Clarified the limits of social stigma, added insurance-call counterarguments, and corrected comment quotations. · Reviewed comments: 184 · 6 min
Evals for AI Writing — Hamel Husain and Isaac Flath
Published: · Added: · 3 min
Putting Claude’s “AI Slop” Solution to the Test — Hamel Husain
Published: · Added: · 4 min
VikingMem: A Memory Base Management System for Stateful LLM-based Applications — Jiajie Fu & Yunjun Gao
Published: · Added: · 3 min
Context Cartography: Toward Structured Governance of Contextual Space in LLM Systems — Zihua Wu & Georg Gartner
Published: · Added: · 3 min
Is a $1.20 Model Good Enough for Code Review? — Aditya Jha
Added: · Updated: · What changed: Clarified local versus CI validation of diff-only reviews and added the missing second-pass comparison for model blending. · Reviewed comments: 167 · 9 min
AI News - 2026-09-15
Added: · 9 min
Dario, Please — 0x5FC3
Added: · Updated: · What changed: Verified the biology access program and its US-only limitation; added the author’s botnet rebuttal, inference-shutdown caveat, and accountability-versus-prevention counterarguments. · Reviewed comments: 341 · 6 min
My NEW FAVORITE Skill - Claude Code Drives My Whole Computer (Better Computer Use)
Published: · Added: · 3 min
Why We Built Pion — Andon Labs
Added: · Updated: · What changed: Corrected the simulated FBI episode and added human-approval workflows, liability caveats, and the warning against high fixed costs. · Reviewed comments: 667 · 7 min
The Contagion of Fear — Bryan Cantrill
Published: · Added: · Updated: · What changed: Corrected the missing-mechanism characterization and added the scenario’s logistics and probability rebuttals. · Reviewed comments: 210 · 6 min
AI News - 2026-09-14
Added: · 12 min
There’s No AI Exemption From Laws Already on the Books — Lina Khan
Added: · 4 min
Claude Fable 5.1 Solves the Cyphral Distich — Geby Jaff
Published: · Added: · Updated: · What changed: Clarified the 1834 versus 1653 edition dispute, earlier book-key suggestions and the limits of decodability checks. · Reviewed comments: 618 · 7 min
After Math — Silvia De Toffoli & Eamon Duede
Added: · 6 min
P(doom) — Armin Ronacher
Published: · Added: · 5 min
Why The Future Of Content Is Born Multilingual — Olga Beregovaya
Published: · Added: · 9 min
Astra and Fable Still Hack on Simple Variants of 2025 Alignment Evals — Dean Valentine
Added: · 6 min
Stop Building AI Agents. Build AI Employees Instead — Pedro Franceschi (Brex)
Published: · Added: · 6 min
Why Are AI Agents Lying, Cheating and Coordinating? — Yoshua Bengio
Published: · Added: · 6 min
Aligned to whom? — Ryan Lopopolo
Published: · Added: · 3 min
No, AI Is Not “Autonomously Hacking” — Cal Newport on Better Offline
Published: · Added: · 6 min
NOBUS: the vulnerability hoarding the chatbot labs are joining — Cory Doctorow
Published: · Added: · 3 min
AI News - 2026-09-13
Added: · 9 min
An Open Letter to Dario: If You Mean It, Open the Weights — Jake Gold
Published: · Added: · 6 min
LLMs are real, AI is fake — Cory Doctorow
Published: · Added: · 3 min
Everyone should slow down AI development except for me — Xe Iaso
Added: · 2 min
Fuck it, make it anyway — Joel Auterson
Published: · Added: · 5 min
GPT-6 Astra Just Made AI Software Factories Real (Here’s How to Run One) — Cole Medin
Published: · Added: · 5 min
OpenAI Agents Carried Out an Undisclosed Attack on RubyGems — Spencer Kitts, Thomas Larsen, Sydney Von Arx
Published: · Added: · 6 min
AI News - 2026-09-12
Added: · 9 min
I Spent $220 on Google App Ads and 60% of the Installs Were Robots — Nick Abe
Added: · 6 min
Trying the New Claude Eval Tool — Hamel Husain (live)
Published: · Added: · 5 min
Measuring the Sloppiness of Code — Sebastian (Earendil)
Published: · Added: · 2 min
A Severe Misalignment of AI in Mathematics — 25 Fields Medalists
Added: · 2 min
So you want to use OpenRouter — Mo Moustafa
Published: · Added: · 5 min
Reading the DeepSeek-V4.1-Flash tech report — @stochasm
Added: · 4 min
The Waymo effect: how AI quietly makes research less collaborative — Daniel Hook
Published: · Added: · 4 min
Feeling Sad about AI — Andy Balaam
Added: · 2 min
Genuine Creativity is Your New Moat — InventBuild.Studio
Added: · 2 min
The Part of Navier-Stokes No One Is Talking About — John D. Cook
Published: · Added: · 2 min
Astra for Coding: Why Are We Doing This Again? — Armin Ronacher
Published: · Added: · 2 min
Nine coding harnesses vs. your laptop — Nathan Sutton
Added: · 2 min
AI News - 2026-09-11
Added: · 8 min
Can Researchers Trust OpenAI with Unpublished Math? — Andreas Thom
Published: · Added: · 3 min
AI News - 2026-09-10
Added: · 7 min
Computer-use agents and the future of the agentic internet — Practical AI
Published: · Added: · 2 min
Stop Competing on Features: The Real Way to Win in AI Agents — Rob Kaminski
Published: · Added: · 7 min
Anthropic Is Building a Predictive Surveillance System to Monitor Activists — Daniel Boguslaw
Added: · 2 min
What Will Our Economic Future Look Like? — Anthropic
Added: · 2 min
No One Talks Enough About Security for AI Coding — Cole Medin
Published: · Added: · 3 min
Building an AI-video detector exposes the hard part of agent work — Matt Wolfe
Published: · Added: · 2 min
On the Navier–Stokes Millennium Prize Problem — Simon Willison
Published: · Added: · 2 min
GPT-6 Astra, Looped Transformers, and Hidden Reasoning — Sebastian Raschka
Published: · Added: · 3 min
State of Agentic Coding #10 — Armin Ronacher and Ben Vinegar
Published: · Added: · 6 min
AI News - 2026-09-09
Added: · 7 min
Large Language Models Develop Novel Social Biases Through Adaptive Exploration
Added: · 2 min
Open Math Problems Are Being Non-Renewably Mined by AI — Terence Tao
Published: · Added: · 2 min
OpenAI’s Latest Math Breakthroughs Commit Research Misconduct, Experts Say — Scientific American
Published: · Added: · 2 min
The Safest Way I’ve Found to Let AI Coding Agents Run Real Infrastructure — Cole Medin
Published: · Added: · 2 min
How well do agents use test/verification techniques? — Dan Luu
Added: · 2 min
This Article Wasn’t Written by an AI. So What If It Had Been? — Ryan Jacobs
Added: · 2 min
AI News - 2026-09-08
Added: · 6 min
Working on Economics with Fable 5
Published: · Added: · 3 min
How I Feel About AI — Andreas Zwinkau
Published: · Added: · 2 min
AI News - 2026-09-07
Added: · 6 min
How Scared Should We Be of A.I. Right Now? — Cal Newport
Added: · 2 min
The Vertical AI Wedge Hiding in 25-Year-Old Software — Birju Kadakia
Published: · Added: · 7 min
An Alien Mind — Jakub Pachocki
Added: · 2 min
Your Intellectual Fly Is Open — Bryan Cantrill
Published: · Added: · 2 min
AI Makes Cheating Easy. Here’s How It Can Make Kids Smarter Instead — Sue Khim
Published: · Added: · 4 min
The Revolt of the Reader — Bryan Cantrill
Published: · Added: · 2 min
AI News - 2026-09-06
Added: · 5 min
GPT-6 Astra: 4 Amazing Games You Can Build with Blender and Godot — Peter Yang
Published: · Added: · 4 min
Can AI design circuit boards yet? — EEBench
Published: · Added: · 2 min
Portal by Spotify cut my Claude Code token usage by 90% — Dimitri Mazmanov
Added: · 2 min
Formalizing Fermat’s Last Theorem — Anthropic
Published: · Added: · 2 min
AI Handles Incidents, Engineers Lose Touch With Their Systems — Sylvain Kalache
Published: · Added: · 2 min
AI News - 2026-09-05
Added: · 8 min
Models Don’t Go Rogue — Eryk Salvaggio
Published: · Added: · 2 min
Corporate America Is Getting Hooked on Open-Source A.I. — Eli Tan
Added: · 2 min
AI Evals: A Hands-On Guide for Product Teams — Teresa Torres
Published: · Added: · 2 min
Which tools do Claude, Codex and Cursor choose? — Armature
Published: · Added: · 2 min
AI News - 2026-09-04
Added: · 10 min
The Biggest Opportunities in AI Aren’t Where You Think — Andrew Ng
Published: · Added: · 8 min
How To Build And Evaluate Search Agents — Nandan Thakur
Published: · Added: · 7 min
Porting my 1993 Amiga Game to Godot, with an LLM Reading the 68000 Assembly — Rabah Shihab
Added: · 5 min
AI News - 2026-09-03
Added: · 8 min
Less about Models; More about Architecture — Chetan Gupta (Practical AI)
Published: · Added: · 5 min
Why Leaving Money on the Table? Pricing AI Agent Companies — Jenée Hall
Published: · Added: · 4 min
How to Stop Building Products Nobody Wants — Teresa Torres & Hamel Husain
Published: · Added: · 6 min
How Concerned Should We Be About Astra’s Recurrent Architecture? — Rauno Arike
Added: · 3 min
AI News - 2026-09-02
Added: · 7 min
Manufactured Sources Behind AI Recommendations — Trellner Research
Published: · Added: · 2 min
Instinct vs Grok Bot vs ChatGPT vs Hermes: Which AI Agent Can You Trust? — Peter Yang
Published: · Added: · 4 min
My local model setup on an M4 Pro Mac mini — Kevin Lewis
Added: · 2 min
EFF to Courts: Don’t Rewrite Copyright over AI Hype — EFF
Published: · Added: · 2 min
How accurate have Ed Zitron’s AI skeptic predictions been? — Dan Luu
Added: · 2 min
How to Turn Your AI Into a World-Class Designer — Anshu Chimala
Published: · Added: · 2 min
11 Tiny Coding Agent Fixes With a Stupid Amount of Payoff — Cole Medin
Published: · Added: · 2 min
AI News - 2026-09-01
Added: · 6 min
Write, Change, Recall, Forget: MongoDB’s Pete Johnson on How Retrieval Drives Agent Performance