61
Trapping misbehaving bots in an AI Labyrinth
How Cloudflare uses generative AI to slow down, confuse, and waste the resources of AI Crawlers and other bots that don’t respect “no crawl” directives.
Most cited
Posts that commenters keep linking to, ranked by how many different people have cited them in Hacker News comments. 593 posts.
62
There's No Fire Alarm for Artificial General Intelligence
What is the function of a fire alarm?
63
Applying BERT models to Search
How new advances in the science of language understanding will help you find more useful information in Search.
64
Agentic Misalignment: How LLMs could be insider threats
New research on simulated blackmail, industrial espionage, and other misaligned behaviors in LLMs.
65
How Claude's text watermarking works
Future Claude models will generate text that contains a watermark to comply with the EU AI Act. We answer questions about how it works and whether it affects outputs.
66
How the voices for ChatGPT were chosen
How the voices for ChatGPT were chosen We worked with industry-leading casting and directing professionals to narrow down over 400 submissions before selecting the 5 voices.
67
Due to DMA, Siri AI delayed in EU for iOS 27 and iPadOS 27
Due to the Digital Markets Act, Apple will not be able to ship Siri AI in the European Union with the release of iOS 27 and iPadOS 27.
68
Combining Machine Learning and Homomorphic Encryption in the Apple Ecosystem
At Apple, we believe privacy is a fundamental human right. Our work to protect user privacy is informed by a set of privacy principles, and…
69
GPT-6 Astra
Introducing GPT-6 Astra, our most intelligent and aligned model yet, with state-of-the-art capabilities across computer use, coding, cybersecurity, and science.
70
ChatGPT Plugins
We’ve implemented initial support for plugins in ChatGPT. Plugins are tools designed specifically for language models with safety as a core principle, and help ChatGPT access up-to-date information, run computations, or use third-party services.
71
OpenAI O3 breakthrough high score on ARC-AGI-PUB
OpenAI o3 scores 75.7% on ARC-AGI public leaderboard.
72
An update on recent Claude Code quality reports
We traced recent reports of Claude Code quality issues to three separate changes. Here's what happened, what we fixed, and what we're changing.
73
Your brain on ChatGPT: Accumulation of cognitive debt when using an AI assistant
This study explores the neural and behavioral consequences of LLM-assisted essay writing. Participants were divided into three groups: LLM, Search Engine, and …
74
March 20 ChatGPT outage: Here’s what happened
An update on our findings, the actions we’ve taken, and technical details of the bug.
75
Subliminal learning: Models transmit behaviors via hidden signals in data
We study subliminal learning , a surprising phenomenon where language models learn traits from model-generated data that is semantically unrelated to those traits. For example, a "student" model learns to prefer owls when trained on sequences of numbers generated by a "teacher" model that prefers owls. This same…
76
Emotion concepts and their function in a large language model
All modern language models sometimes act like they have emotions. What’s behind these behaviors? Our interpretability team investigates.
77
The Illusion of Thinking: Strengths and Limitations of Reasoning Models
Recent generations of frontier language models have introduced Large Reasoning Models (LRMs) that generate detailed thinking processes…
78
Attacking Machine Learning with Adversarial Examples
Stay up to speed on the rapid advancement of AI technology and the benefits it offers to humanity.
79
Task-Completion Time Horizons of Frontier AI Models
Our measurements of the time horizons for public frontier language models.
80
Federated Learning: Collaborative Machine Learning without Centralized Training
Posted by Brendan McMahan and Daniel Ramage, Research ScientistsStandard machine learning approaches require centralizing the training data on one ...
81
Code execution with MCP for efficient agents
Learn how code execution with the Model Context Protocol enables agents to handle more tools while using fewer tokens, reducing context overhead by up to 98.7%.
82
OpenAI's GPT-3 Language Model: A Technical Overview
Chuan Li, PhD reviews GPT-3, the new NLP model from OpenAI. The technical overview covers how GPT-3 was trained, GPT-2 vs. GPT-3, and GPT-3 performance.
83
My AI skeptic friends are all nuts
My smartest friends have bananas arguments about LLM coding.
84
An OpenAI model has disproved a central conjecture in discrete geometry
An OpenAI model solved the 80-year-old unit distance problem, disproving a major conjecture in discrete geometry and marking a milestone in AI-driven mathematics.
85
The path to ubiquitous AI (17k tokens/sec)
By Ljubisa Bajic Many believe AI is the real deal. In narrow domains, it already surpasses human performance. Used well, it is an unprecedented amplifier of human ingenuity and productivity. Its widespread adoption is hindered by two key barriers: high latency and astronomical cost. Interactions with language models…
86
Google supercharges machine learning tasks with TPU custom chip
Machine learning provides the underlying oomph to many of Google’s most-loved applications. In fact, more than 100 teams are currently using machine learning at Google today, from Street View, to Inbox Smart Reply, to voice search.
87
OpenAI Five
Stay up to speed on the rapid advancement of AI technology and the benefits it offers to humanity.
88
Hardening Firefox with Anthropic's Red Team
Claude Opus 4.6 found 22 Firefox vulnerabilities in two weeks, 14 rated high-severity, with proofs-of-concept and candidate patches for Mozilla.
89
Sycophancy in GPT-4o
We have rolled back last week’s GPT‑4o update in ChatGPT so people are now using an earlier version with more balanced behavior. The update we removed was overly flattering or agreeable—often described as sycophantic.
90
Differentiable Neural Computers
In a recent study in Nature, we introduce a form of memory-augmented neural network called a differentiable neural computer, and show that it can learn to use its memory to answer questions about complex, structured data, including artificially generated stories, family trees, and even a map of the London Underground…