Full-text search across 17414 articles. Combine with topic and date filters; results sorted by relevance.
Akamai Technologies reports that inference is replacing training as the dominant AI workload, requiring a shift from hyperscale datacenters to distributed, power-aligned commercial scale facilities to meet sub-500 millis…
Inna Braverman, founder and CEO of Eco Wave Power, a Nasdaq-listed renewable energy company, said the company's wave energy technology can help meet growing electricity demand from AI infrastructure and data centers. The…
Moonshot AI released the weights for Kimi K3 on July 27, a 2.8-trillion-parameter Mixture-of-Experts model with a 1-million-token context window, scoring third globally behind Claude Fable 5 Max and GPT-5.6 Sol Max on GD…
Mozilla's llamafile v0.10.5 release adds support for two large local models: the 6GB Ternary Bonsai 27B and Poolside's 118B Laguna-S-2.1 coding MoE, by syncing with a more recent llama.cpp core. The update also ships pre…
A developer built a personal AI-powered wiki using Obsidian, the Karpathy LLM Wiki plugin, and Ollama, processing over 330 notes locally after spending $10 per day for three days on public API costs. The setup runs a Qwe…
Protocol Labs, Bluesky, number zero, and Witchcraft Software have published the User Controlled Authorization Network (UCAN) specification, a trustless, secure, local-first, user-originated, distributed authorization sch…
MIRI and other AI governance researchers argue that society should not wait for a 'warning shot' to address existential risks from superintelligent AI, as their analysis of potential warning shots—AI-caused pandemics, mi…
A developer trained a small language model from scratch on an $8 ESP32-S3 microcontroller, demonstrating that training, not just inference, is possible on ultra-low-cost hardware. The project, inspired by the esp32-ai re…
OpenAI publicly rebutted Apple's trade secret lawsuit, asserting that Apple's claims are baseless and that Apple's own counsel confused an employee's identity and never contacted OpenAI's legal team before filing. OpenAI…
A developer is seeking contributors for Zentrail IDE, an open-source, AI-native desktop IDE designed to enable multiple AI agents to collaborate with developers in a single workspace. The project is in early architecture…
Hubmesh, a new Python library, improves multi-hop RAG retrieval by adding a centrality-aware planner over existing vector databases, eliminating LLM calls in the query path. The library uses multi-component seed selectio…
Cursor released an open-source megakernel that delivers a 41% increase in tokens per second, potentially saving billions of dollars at scale, according to a post by Stuart Sul, who leads the team behind the project and c…
Friedrich Hayek's 1945 'knowledge problem' remains unresolved for AI because large language models lack continual learning, limiting their economic value in firm-specific tasks, according to a Substack essay by CP Walker…
Vercel has launched a new feature allowing users to search for and purchase domains directly from its dashboard, with automatic nameserver configuration for projects. The feature is accessible via the Buy button in domai…
Retrieval-Augmented Generation (RAG) is the key to building AI agents that don't hallucinate, according to a new guide that outlines how to set up retrieval layers using vector stores like Chroma, FAISS, Qdrant, and Pine…
DeepSeek's V4 Flash 0731, a 284B-parameter Mixture-of-Experts model with 13B active parameters and a native 1M-token context under an MIT license, is cheaper to use via the API than to run locally, according to a reviewe…
A letter to the editor by Jack Fleck of Oakland criticizes California's Proposition 45, arguing it would minimize environmental review for data centers, freeways, fossil-fuel power plants, dams, and sprawling land uses, …
Paul Christiano has returned to the Alignment Research Center (ARC) as executive director, focusing on mechanistic interpretability and misalignment detection. ARC is hiring researchers, a chief of staff, and an automati…
A two-week benchmark of ten AI translation tools found Claude 3.5 Sonnet, GPT-4 Turbo, and DeepL Pro best for developer use cases, with Claude achieving 97% accuracy on technical content versus 94% for GPT-4 Turbo and 89…
Telegram founder Pavel Durov accused organized extortionists of exploiting Apple's App Store review process after Apple temporarily removed Telegram from the iOS marketplace over hidden AI-modified child sexual abuse mat…