781
AI Canon
A curated list of resources we’ve relied on to get smarter about modern AI, including generative AI, LLMs, and transformer models.
Topics /
AI & Machine Learning
The most popular blog posts about ai & machine learning on Hacker News. 6,306 posts.
782
AI is killing B2B SaaS
SaaS is the most profitable business model on Earth.1 It’s easy to understand why: build once, sell the same thing again ad infinitum, and don’t suffer any marginal costs on more sales. I have been writing software for more than half my life. In the last year itself, I’ve talked to hundreds of founders and operators…
783
Zed now predicts your next edit with Zeta, our new open model
From the Zed Blog: A tool that predicts your next move. Powered by Zeta, our new open-source, open-data language model.
784
Everything wrong with MCP
Explaining the Model Context Protocol and everything that might go wrong.
785
RAG Is Simpler Than You Think
Six approaches to retrieval-based AI, from minimal to elaborate
786
Are OpenAI and Anthropic losing money on inference?
Deconstructing the real costs of running AI inference at scale. My napkin math suggests the economics might be far more profitable than commonly claimed.
787
Local, CPU-Friendly, High-Quality TTS (Text-to-Speech) with Kokoro
Just a few years ago, realistic local speech generation seemed unimaginable. Today, its quality is exceptional and, crucially, it delivers these results without compromising privacy.
788
Llama from scratch, or how to implement a paper without crying
Llama from scratch I want to provide some tips from my experience implementing a paper. I'm going to cover my tips so far from implementing a dramatically...
789
Not everyone is using AI for everything
People are consuming AI like they eat meat: some are embracing it, some are limiting their use of it, and some are avoiding it altogether.
790
Higher usage limits for Claude and a compute deal with SpaceX
We’ve raised Claude's usage limits and agreed a new compute partnership with SpaceX that will substantially increase our capacity in the near term.
791
Grok Code Fast 1
We're thrilled to introduce grok-code-fast-1, a speedy and economical reasoning model that excels at agentic coding.
792
An LLM is a lossy encyclopedia
Since I love collecting questionable analogies for LLMs, here's a new one I just came up with: an LLM is a lossy encyclopedia. They have a huge array of facts …
793
Lessons after a Half-billion GPT Tokens
[custom_views_and_hn_comments hn_id=”40015185″] My startup Truss (gettruss.io) released a few LLM-heavy features in the last six months, and the narrative around LLMs that I read on Hacker News is now starting to diverge from my reality, so I thought I’d share some of the more “surprising” lessons after churning…
794
795
The FBI Is Tracking Our Faces in Secret, and We’re Suing
There can be no accountability if there is no transparency.
796
How OpenAI delivers low-latency voice AI at scale
How OpenAI rebuilt its WebRTC stack to power real-time Voice AI with low latency, global scale, and seamless conversational turn-taking.
797
CICERO: An AI agent that negotiates, persuades, and cooperates with people
Meta AI has built CICERO, the first AI to achieve human-level performance in the popular strategy game Diplomacy. It's a breakthrough toward building AI that can use language to work with people to achieve strategic goals.
798
AI is stifling new tech adoption?
AI coding assistants are React evangelists.
799
Deep Learning Is Applied Topology
Everything lives on a manifold
800
Prompt injection explained, with video, slides, and a transcript
I participated in a webinar this morning about prompt injection, organized by LangChain and hosted by Harrison Chase, with Willem Pienaar, Kojin Oshiba (Robust Intelligence), and Jonathan Cohen and Christopher …
801
Mistral ships Le Chat – enterprise AI assistant that can run on prem
The most powerful AI platform for enterprises. Customize, fine-tune, and deploy AI assistants, autonomous agents, and multimodal AI with open models.
802
Quantized Llama models with increased speed and a reduced memory footprint
As our first quantized models in this Llama category, these instruction-tuned models retain the quality and safety of the original 1B and 3B models, while achieving 2-4x speedup.
803
Visualize data instantly with machine learning in Google Sheets
Today we’re rolling out new features in Sheets that make it even easier for you to visualize and share your data, and find insights your teams can act on.
804
Over fifty new hallucinations in ICLR 2026 submissions
GPTZero used our Hallucination Check tool to find 50+ hallucinations under review at ICLR, each of which were missed by 3-5 peer reviewers.
805
How to setup a local coding agent on macOS
Running Gemma 4 26B-A4B and Qwen3.6 35B-A3B locally with llama.cpp, MTP speculative decoding, multimodal support, and PI as a coding agent.
806
Kimi K3 Architecture Overview and Notes
Short architecture note on Kimi K3, including LatentMoE, Kimi Delta Attention, Attention Residuals, NoPE, multimodality, and inference-efficiency choices.
807
Drowning in AI Generated Garbage: the silent war we are fighting
Drowning in AI Generated Garbage : the silent war we are fighting par Ploum - Lionel Dricot.
808
How we hacked McKinsey's AI platform
An autonomous AI agent found a SQL injection in McKinsey's Lilli AI platform. What it extracted was worse than we expected.
809
Disagreement among frontier LLMs on real-world fact-checks
63% of real-world claims expose disagreement among the five top frontier LLMs. Methodology, breakdowns, and data CSV.
810
Driverless robotaxi fleet paralyzed for hours in San Francisco
Apparently, the first system breakdown of a driverless robot taxi fleet occurred. In San Francisco, at least a dozen autonomous Chevrolet Bolts from GM Cruise Automation were spotted blocking the i…