Skip to content
AI NEWS JOURNAL
AI, IN PERSPECTIVE

Search AI news

Search headlines, summaries and articles to connect the dots.

Newest first

Fluorescent Thanksgiving

I looked at a bunch of Thanksgiving foods under UV light and they basically all glow. Don't worry, it's fine!

Show HN: Lightweight FUSE-based encrypted, compressed filesystem for Linux

A personal project for digging deeper into Linux system programming, ObFUSEcate implements a file backed virtual block device that automatically performs compression and encryption as blocks are accessed. This virtual device can be mounted via FUSE (requiring no root privileges) to provide, effectively, a "live" compressed, encrypted archive.Written in C++, and intended to have minimal dependencies, it has custom 32 bit arithmetic encoding/decoding, ChaCha20-Poly1305 authenticated encryption, and CRC8 + XOR parity error-correction. Comments URL: https://news.ycombinator.com/item?id=50585222 Points: 1 # Comments: 0

Amazon SageMaker Inference: 2026 launches in review, part 2

This second installment reviews 13 more Amazon SageMaker AI inference launches from 2026, spanning configuration recommendations, fitting more models onto each GPU, serving models too large for one instance, cross-Region routing, and post-launch tuning, bringing the year's total to 26.

Introducing enhanced infrastructure observability in Amazon SageMaker HyperPod

Amazon SageMaker HyperPod now offers enhanced observability to help you diagnose GPU training issues faster. It brings together diagnostic and profiling capabilities for training workloads on both Amazon EKS and Slurm orchestrators, pinpointing issues across GPU, network, and host layers.

Genie 4 is now available

Google introduces a new Project Genie experience for eligible Google AI Ultra subscribers, featuring Genie 4—the most realistic and interactive world model yet—enabling real-time prompt-based changes to environments, native audio adaptation, and extended session capabilities. Upgrades to Genie 3 include longer runs, 360-degree panorama grounding, and real-time in-world events. The Project Genie agent allows users to create interactive worlds through natural language prompts, while the community gallery offers curated and user-generated worlds for exploration and remixing. Safety measures like SynthID watermarking and content screening are in place, and ongoing research addresses challenges such as camera latency and scene permanence.

Introducing the Amazon SageMaker JumpStart Model Leaderboard

The Amazon SageMaker JumpStart Model Leaderboard compares 70+ foundation models side by side on quality, speed, and cost, directly in SageMaker Studio. Learn what the leaderboard shows, how AWS benchmarks models, and how to go from a shortlist to a deployed model.

180B Parameters, 32GB of RAM: Pushing MoE Inference to the Limit with Transformers

The author explores streaming MoE (Mixture of Experts) model experts directly from SSD to achieve near-Strata performance on a 32GB MacBook M5, using a custom implementation in Hugging Face Transformers. By optimizing expert caching, leveraging Metal kernels, and minimizing host-GPU synchronization delays, the approach reaches ~14 tok/s—within 6% of Strata's performance—demonstrating that efficient expert streaming is feasible on consumer hardware without C++ extensions.

Govern and discover records with common metadata in AWS Agent Registry

Custom metadata schemas in AWS Agent Registry let an administrator define typed, validated fields that every agent, MCP server, and skill must carry. This post follows an MCP record from schema definition through registration, enrichment, approval, and discovery, so teams govern and find resources with one shared set of facts.

Reducing token costs and latency in Amazon Bedrock AgentCore JSON workloads

As production agents on Amazon Bedrock AgentCore process multi-megabyte JSON payloads, token costs and latency climb. This post shares three composable strategies: offloading heavy artifacts to Amazon S3, on-demand result paging, and dictionary compaction. Together they cut per-turn token use by up to 71% and latency by up to 80%.

The Silent Consensus

Dean Ball, Head of Strategic Futures at OpenAI, examines how more capable AI could reshape society and how people can retain control of their institutions.

China’s AI-for-Science Boom Is Selling Speed, Not Discovery

Investors have poured more than 20 billion yuan into Chinese firms that pair models with lab automation. Cycle times for early molecular work are falling. Paying customers still buy faster screens and cleaner pipelines, not machines that invent science on their own.

How the World’s Most Advanced AI Infrastructure Supply Chain Invents the Future

Deb Shoquist leads operations at NVIDIA, overseeing the supply chain and manufacturing behind its AI platforms. That description sums up the job, but it doesn’t begin to encompass what Shoquist has made possible over decades. Shoquist, her team and their global network of manufacturing partners enable NVIDIA to introduce world-leading technologies year-round. Much of the […]

Novo Nordisk modernizes manufacturing process documents with Amazon Bedrock AgentCore

Novo Nordisk's Manufacturing Technology team within the Tablet area team built a process-intelligence assistant on Amazon Bedrock AgentCore, using a GraphRAG architecture that pairs a knowledge graph in Amazon Neptune with hybrid search in Amazon OpenSearch Serverless to turn millions of controlled manufacturing documents into grounded, cited answers.

NVIDIA Sweeps MLPerf Training and Endpoints as Vera Rubin Debuts With Leading Performance

In its MLPerf benchmarks debut, with results released today, the NVIDIA Vera Rubin platform delivers leading performance across training, post-training and agentic inference. NVIDIA swept every MLPerf Training v6.1 benchmark and, in its first MLPerf Endpoints submission, delivered more than 10x generational gains in throughput per watt for agentic AI — measuring how much agentic […]

Cheap AI Tracks Are Crowding Human Arrangers Out of China’s Playlist Economy

Chinese music platforms now add more than 100,000 songs a day, and a rising share is machine-made. Clients who once paid thousands for an arrangement now shop for a few hundred yuan and a menu of AI drafts. The middle of the profession is getting squeezed; the top and the pure commodity layers are sorting out differently.

LLMON: An LLM-Native Markup Language, now in Granite 4.1 3B and 8B

IBM has introduced LLMON (LLM Object Notation), a new markup language designed to explicitly communicate structure and semantics to large language models. Built to address issues like prompt injection attacks and brittle behavior in smaller models, LLMON uses reserved special-token delimiters to mark semantically coherent components such as instructions, data, and requirements, enabling precise binding between them. Two experimental preview models—granite-4.1-3b-LLMON-preview and granite-4.1-8b-LLMON-preview—are now available, alongside a GitHub repository with conversion tools, workflows, and examples. LLMON enhances robustness by separating instruction from data, reducing indirect prompt injection success rates nearly to zero while maintaining performance comparable to standard Granite 4.1 Instruct models. It also supports custom schemas via LoRA fine-tuning, as demonstrated in ScienceWorld and tool-calling agent use cases. The approach is distinct from JSON, which lacks model-friendly structure, and complements existing chat templates by adding semantic clarity across roles.

Our first phase of embedded assessments

OpenAI begins embedded assessments with Apollo Research, METR, and Transluce, and Goodfire to independently evaluate model alignment, safeguards, and monitoring systems.