811
LLMs work best when the user defines their acceptance criteria first
One of the simplest tests you can run on a database:
Most cited
Posts that commenters keep linking to, ranked by how many different people have cited them in Hacker News comments. 1,015 posts.
812
What can LLMs never do?
On goal drift and lower reliability. Or, why can't LLMs play Conway's Game Of Life?
813
OpenAI and Microsoft extend partnership
We’re happy to announce that OpenAI and Microsoft are extending our partnership.
814
Unrolling the Codex agent loop
A technical deep dive into the Codex agent loop, explaining how Codex CLI orchestrates models, tools, prompts, and performance using the Responses API.
815
Ironwood: The first Google TPU for the age of inference
We’re introducing Ironwood, our seventh-generation Tensor Processing Unit (TPU) designed to power the age of generative AI inference.
816
Mozilla.ai: Investing in Trustworthy AI
We’re committing $30M to build Mozilla.ai: A startup — and a community — building a trustworthy, independent, and open-source AI ecosystem. We’re o
817
818
Gemma 3n preview: Mobile-first AI
Introducing Gemma 3n – the latest Google open model for accessible AI, featuring unique flexibility, privacy, and expanded multimodal capabilities on mobile devices.
819
820
Patterns for building LLM-based systems and products
Evals, RAG, fine-tuning, caching, guardrails, defensive UX, and collecting user feedback.
821
Superpowers: How I'm using coding agents in October 2025
I'm Jesse. I make stuff. Software, hardware. Very occasionally, trouble.
822
My Lethal Trifecta talk at the Bay Area AI Security Meetup
I gave a talk on Wednesday at the Bay Area AI Security Meetup about prompt injection, the lethal trifecta and the challenges of securing systems that use MCP. It wasn’t …
823
Microsoft demos 'holoportation' 3D presence tech with HoloLens
Holoportation is a new type of 3D capture technology that allows high-quality 3D models of people to be reconstructed, compressed and transmitted anywhere in the world in real time. When combined with mixed reality displays such as HoloLens, this technology allows users to see, hear, and interact with remote…
824
Kimi K3, and what we can still learn from the pelican benchmark
Chinese AI lab Moonshot AI announced Kimi K3 this morning, describing it as their “most capable model to date, with 2.8 trillion parameters”. It’s currently available via their website and …
825
There are no "rogue" AI agents
Identifying them as such only lets companies like OpenAI off the hook.
826
Decisions API is in public beta
Use the Decisions API to check conditions, select from fixed options, and score text, images, and JSON against a rubric.
827
YOLOv5: State-of-the-art object detection at 140 FPS
YOLOv5 license and commercial-use guide: AGPL-3.0 terms, specs, resources, and alternatives. Compare YOLOv5 against other vision models free.
828
Duolingo Max, a learning experience powered by GPT-4
Duolingo’s newest subscription, Duolingo Max, offers a powerful AI-backed learning experience.
829
Writing code is cheap now
Writing code is cheap now - Agentic Engineering Patterns
830
NaturalSpeech: End-to-end text to speech synthesis with human-level quality
Text to speech (TTS) has made rapid progress in both academia and industry in recent years. Some questions naturally arise that whether a TTS system can achieve human-level quality, how to define/judge that quality and how to achieve it. In this paper, we answer these questions by first defining the human-level…
831
LLMs aren't world models
I believe that language models aren’t world models. It’s a weak claim — I’m not saying they’re useless, or that we’re done milking them. It’s also a fuzzy-sounding claim — with its trillion weights, who can prove that there’s something an LLM isn't a model of? But I hope to make my claim clear and persuasive enough…
832
The Generative AI Con
It's been just over two years and two months since ChatGPT launched, and in that time we've seen Large Language Models (LLMs) blossom from a novel concept into one of the most craven cons of the 21st century — a cynical bubble inflated by OpenAI CEO Sam Altman built
833
The next chapter of the Microsoft–OpenAI partnership
Microsoft and OpenAI sign a new agreement that strengthens its long-term partnership, expands innovation, and ensures responsible AI progress.
834
835
Images that fool computer vision raise security concerns
Computers are learning to recognize objects with near-human ability. But Cornell researchers have found that computers, like humans, can be fooled by optical illusions, which raises security concerns.
836
Image Kernels explained visually
An image kernel is a small matrix used to apply effects like the ones you might find in Photoshop or Gimp, such as blurring, sharpening, outlining or embossing. They're also used in machine learning for 'feature extraction', a technique for determining the most important portions of an image. In this context the…
837
838
Prompt engineering vs. blind prompting
"Prompt Engineering" emerged from the growth of language models to describe the process of applying prompting to effectively extract information from language models, typically for use in real-world applications.
839
Copilot Internals
Hacky repo to see what the Copilot extension sends to the server
840
Using machine learning to recreate photorealistic portraits of Roman Emperors
Artistic interpretations of the 54 Roman emperors of the Principate, created with AI and historical references.