Full-text search across 16983 articles. Combine with topic and date filters; results sorted by relevance.
The MARIN community is publicly training a 535B-parameter mixture-of-experts large language model with 23B active parameters on 18 trillion tokens, streaming the process live via Weights & Biases. The 'Hero Run' is part …
Hugging Face forum user sebastianchris posted an explanation on August 21, 2026, detailing how tokens function in generative AI, from text input to tokenization to intelligent answer generation. The post aims to demystif…
Gwyneth Paltrow's attempt to deflect criticism over her hosting a Hamptons dinner for OpenAI CEO Sam Altman by posting satirical invitations backfired, drawing further backlash from the creative community. The August 29t…
In the August edition of a developer community update, a developer shared progress on ZVector, a vector database engine built in Zig, focusing on compute and memory performance using @Vector primitives and cache-line ali…
Reddit Inc., the social news platform, faces an uncertain future as artificial intelligence tools increasingly aggregate and summarize its user-generated content, potentially reducing direct traffic to the site. The Econ…
Protesters brought a guillotine to a city council meeting to oppose a proposed AI data center, according to reports. The demonstration highlights community concerns over the project's potential impacts.
A new five-level ladder for individual AI proficiency, ranging from L0 'New' to L5 'Loop', has been proposed by an unnamed source seeking feedback from the Hacker News community. The levels describe escalating capabiliti…
A Hacker News user asks whether AI tools like Cowork and Codex are replacing spreadsheets as the go-to for simple apps, noting their own spreadsheet use has dropped immensely since adopting these tools. The post, which h…
Developers report that LLM-powered chatbots still struggle to maintain coherent conversations over multiple turns, with issues such as memory loss and context confusion emerging in longer interactions. A Reddit user aske…
The Debian project is holding a vote on eight options regarding the use of large language models (LLMs) in its development processes, as reported by LWN.net. The vote reflects ongoing community debate about integrating A…
A developer sparked a debate on Hacker News with a mock-up of a GUI for AI agents, asking the community for feedback on layout and interaction design. Contributors favored node-based canvases, timeline views, and form-dr…
A new website, Slop Or Not, lets users vote on whether AI-generated content is 'slop,' with early results showing 78% of users deem Cloud Scanner of Death, a professional-grade cloud security assessment tool, as not slop…
A bipartisan backlash against data centers is emerging in the U.S., with Wisconsin Republican gubernatorial candidate Tim Tiffany attacking Democratic rival David Crowley in a new ad for supporting data center tax breaks…
Technical Artist Ariescar (Gobkit) released AnyCreature, an open-source, mesh-based generative harness for 3D creature creation, featuring three automated pipeline checks: QC thresholds for game-engine specs, quantified …
A developer argues that claims of massive productivity gains from LLM-assisted programming are misleading because LLMs are non-deterministic, and combining two variable systems (human and LLM) compounds variance, making …
A Hacker News user asked the community how much AI they use and what they consider excessive, noting that friends who criticize AI often cite water usage while themselves consuming meat and new clothes, and linked to a B…
A Yahoo Finance report finds that consumer opposition to AI data centers is growing, with critics calling it a 'hot Luddite summer' as communities push back against the infrastructure's energy and water demands. Senior R…
An engineer's guide to MCP servers for Figma, Jira, GitLab, and Playwright reveals that Figma offers both local and remote servers, while Atlassian provides no separate Jira server—only a unified server covering Jira, Co…
A forum user on Level1Techs detailed a new LLM inference workstation built with four PNY NVIDIA RTX PRO 6000 Blackwell GPUs (384GB VRAM), an AMD Ryzen Threadripper Pro 9975WX CPU, an ASUS Pro WS WRX90E-SAGE SE motherboar…
A developer's benchmark of coding agents on 108 bugs found that models memorized answers, raising questions about evaluation efficacy. The Signal Bench, created by darvh, showed high scores but flagged memorization, prom…