Skip to content
gentic.news — AI News Intelligence Platform
Connecting to the Living Graph…
gentic.news

Latest AI News

Every story, newest first — research, funding, product launches, policy, and analysis.

Google VP Robby Stein discussing AI-native search costs and volume growth in a tech conference setting with data…
Products & Launches
85

Google Search VP: AI-native search costs more, grows volume

Google VP Robby Stein says AI-native Search costs more but grows volume. AI Mode breaks queries into sub-searches.

x.com/6d ago/3 min read
infrastructureai searchgoogle
3D wireframe model of a chair overlaid with depth estimation heatmap colors, with small icons for surface normal and…
AI Research
91

SpatialBench: New Benchmark Tests Foundation Models on 3D Tasks

SpatialBench, a new benchmark from ropedia_ai, evaluates spatial foundation models across 7 tasks and 5 datasets, testing depth estimation, surface normal prediction, and 3D object detection.

x.com/6d ago/3 min read/Multi-Source
foundation-modelsbenchmarkscomputer-vision
A person holding a smartphone with a bounding box around a cat on the screen, illustrating real-time object detection
AI Research
85

RF-DETR Hits Hugging Face Transformers: SOTA Real-Time Detection

Roboflow's RF-DETR, a SOTA real-time detection model, integrated into Hugging Face Transformers, bridging DETR accuracy with real-time speed.

x.com/6d ago/3 min read
open-sourcereal-time-inferenceroboflow
Anthropic engineer presenting a diagram of dynamic agent permissions, showing contextual access controls scaling…
Products & Launches
75

Anthropic: Agent Permissions Should Evolve with Capability

Anthropic advocates dynamic agent permissions. The blog proposes contextual controls as agents learn, mirroring human access evolution.

x.com/6d ago/3 min read
ai safetyagent systemsenterprise ai
NVIDIA Vera CPU on a test bench with cooling hardware, surrounded by cables and monitoring equipment in a lab setting
AI Research
100

NVIDIA Vera CPU Benchmarks: 1.55x Faster Than Intel Xeon in Phoronix Tests

NVIDIA Vera CPU benchmarks show 1.55x performance over Intel Xeon 6980P and 10% over AMD EPYC 9575F, with 1.2 TB/s memory bandwidth.

x.com/6d ago/3 min read/Widely Reported
hardwarebenchmarksnvidia

Microsoft RAMPART Brings Pytest-Based Safety Testing to A…

Products & Launches
89

Microsoft RAMPART Brings Pytest-Based Safety Testing to AI Agents

Microsoft's RAMPART brings pytest-native safety testing to AI agents, covering adversarial attacks and benign failures, addressing a critical gap in agent development.

x.com/6d ago/3 min read
ai safetymicrosofttesting
A hyperrealistic 3D-rendered human face with intricate skin pores and micro-details, illuminated by studio lighting…
AI Research
85

DeemosTech Rodin Gen-2.5: 10M-Polygon 3D GenAI in 4 Seconds

DeemosTech claims Rodin Gen-2.5 generates 10M polygon 3D models in 4 seconds with skin microstructures, but provides no benchmarks or technical details.

x.com/6d ago/3 min read
deemostechgenerative models3d generation
Anthropic engineers discuss code on monitors in a modern office, with a diagram showing AI agent permission levels…
AI Research
94

Anthropic Sandboxing Agents by Capability Level

Anthropic sandboxes agents by capability level, limiting destructive actions as agents gain autonomy in Claude.

x.com/May 26, 2026/3 min read/Multi-Source
access controlagent frameworksanthropic

MiniMax M3 Sparse Attention: 15.6x Decoding Speedup at 1M…

AI Research
100

MiniMax M3 Sparse Attention: 15.6x Decoding Speedup at 1M Tokens

MiniMax M3 sparse attention achieves 9.7x prefilling and 15.6x decoding speedup at 1M tokens, reversing M2's full-attention stance.

x.com/May 26, 2026/3 min read/Widely Reported
open sourcemachine learningai
A diagram showing a knowledge graph with connected nodes representing entities and relationships, with a central…
AI Research
95

Zep AI's Graphiti: Agent Memory Without Schema Is Just Storage

Zep AI's Graphiti enforces Pydantic schemas on LLM entity extraction, preventing generic label collapse and enabling precise querying of agent memory.

x.com/May 26, 2026/3 min read/Widely Reported
knowledge graphsai infrastructureopen-source
A person typing on a laptop, code visible on screen, surrounded by AI-related icons and gears, representing…
Opinion & Analysis
72

No Rigorous Productivity Tests Exist for Post-2025 Autonomous Coding Tools

No productivity studies exist for autonomous coding tools launched December 2025. All research predates the Claude Code/Codex revolution, creating a major knowledge gap.

x.com/May 26, 2026/3 min read
productivity researchai codingautonomous agents
A diagram showing an AI agent with code as its central reasoning layer, surrounded by a harness linking to tools and…
AI Research
89

Meta-Stanford Survey: Code as Agent Harness Improves AI Reasoning

Meta, Stanford, Illinois survey argues AI agents work better with code as their main working layer, calling it an agent harness.

x.com/May 25, 2026/3 min read
code generationsurvey paperai agents
Apple engineer working on Siri code on a MacBook, with a glowing AI model diagram on screen showing 1.2T parameters
Products & Launches
85

Apple Using Custom 1.2T-Parameter Google Model for Siri Overhaul

Apple using custom 1.2T-parameter Google model for Siri, per Reuters. Model larger than Gemini 3.5 Flash's 300B parameters; simple queries run locally.

x.com/May 25, 2026/3 min read
siriappleai models
A sleek AI agent interface displays a complex workflow diagram with interconnected skill nodes, while a bar chart in…
AI Research
89

Microsoft SkillOpt Trains Agent Skills in Text Space, Beats 52/52 Benchmarks

Microsoft's SkillOpt trains agent skills in text space, achieving best or tied-best results in all 52 settings across 6 benchmarks and 7 models.

x.com/May 25, 2026/3 min read
benchmarksmicrosoftagent-training
A split-screen demo showing a webpage URL pasted on the left and Claude AI regenerating the site's exact colors…
Products & Launches
80

Claude Copies Any UI From URL: Colors, Fonts, Layout

Claude can now copy any UI from a URL, extracting colors, fonts, and layout to rebuild the interface, per a demo by @HowToAI_. Anthropic has not officially confirmed the feature.

x.com/May 25, 2026/3 min read
anthropicfrontend developmentai tools

Alibaba + Nanjing Univ Claim 9.36X Faster Million-Token P…

AI Research
85

Alibaba + Nanjing Univ Claim 9.36X Faster Million-Token Prefill vs FlashAttention-2

Alibaba + Nanjing Univ claim 9.36X faster million-token prefill vs FlashAttention-2, targeting the key bottleneck in long-context LLM inference.

x.com/May 25, 2026/3 min read
efficiencyllm inferenceattention mechanisms
Huawei engineer presenting the Tau Scaling Law on a large screen at IEEE ISCAS 2026, showing transistor density…
AI Research
99

Huawei's τ Scaling Law Redefines Transistor Race Without EUV

Huawei's τ Scaling Law at IEEE ISCAS replaces geometric transistor scaling with time-based optimization, targeting 1.4nm density by 2031 without EUV, challenging US export controls.

x.com/May 25, 2026/3 min read
ai-hardwarechinatrade-war
A GitHub repository page showing Claude.md with 152K stars, while a developer observes LLM failure patterns on a…
Products & Launches
77

Claude.md Hits 152K GitHub Stars; Karpathy Notes LLM Failure Patterns

Claude.md hits 152K GitHub stars. Karpathy notes LLMs fail consistently, driving demand for standardized prompt templates.

x.com/May 25, 2026/3 min read
open sourceanthropicai tools
A hooded figure types on a laptop in a dark room, with glowing lines of code and package manager logos for npm…
Products & Launches
85

TrapDoor supply-chain attack hits npm, PyPI, Crates.io — weaponizes AI config files

TrapDoor planted 34 malicious packages on npm, PyPI, and Crates.io, and injected poisoned AI config files into repos to weaponize Claude Code and Cursor.

x.com/May 24, 2026/3 min read
open-sourcemalwarecybersecurity
NVIDIA CEO Jensen Huang speaking at a tech conference, gesturing with his hand while standing on stage in front of a…
Opinion & Analysis
77

Jensen Huang Wants Zero Coding at NVIDIA — 'Purpose vs Task'

Jensen Huang wants zero coding by NVIDIA engineers, framing it as a task to minimize. The bet is AI-generated code will match human output for performance-critical software.

x.com/May 24, 2026/3 min read
automationsoftware engineeringai

Hermes Agent Desktop App Launches for Multi-Agent Management

Products & Launches
86

Hermes Agent Desktop App Launches for Multi-Agent Management

Hermes Agent launched a desktop app for orchestrating autonomous AI agents with persistent memory and continuous workflows, announced via X.

x.com/May 24, 2026/3 min read
agent orchestrationdesktop softwareai agents
A data center rack with Ascend 910B NPUs illuminated by blue LEDs, beside a graph showing VRAM savings from 1.58-bit…
AI Research
87

ModelBest Drops BitCPM-CANN: First 1.58-bit LLM on Ascend 910B

ModelBest released BitCPM-CANN, the first 1.58-bit ternary LLM on Ascend 910B NPUs, using 6× less VRAM than BF16 with minimal capability loss.

x.com/May 24, 2026/3 min read
hardwarequantizationllm
A modern WiFi router on a desk with glowing LEDs, surrounded by abstract wireless signal waves, symbolizing the…
AI Research
75

WiFi routers can identify individuals with near-perfect accuracy, KIT shows

KIT researchers show WiFi routers can identify individuals with near-perfect accuracy via beamforming feedback, tested on 197 subjects.

x.com/May 24, 2026/3 min read
privacywifisecurity
A dramatic biblical scene with a young man in ancient armor standing before a giant warrior, surrounded by a rocky…
Products & Launches
85

Kling AI Video Enters Hollywood Production with 'House of David'

Kling AI video used in 'House of David', first Hollywood production at industrial scale. Show reached 44M+ viewers, #1 on Prime Video U.S.

x.com/May 24, 2026/3 min read
amazonkuaishouai video
A researcher at a computer screen in a modern university library, with the PaperDebugger interface visible…
AI Research
78

PaperDebugger Open-Sourced: NUS Tool Auto-Fixes Academic Writing

NUS open-sourced PaperDebugger, an in-editor tool that auto-fixes academic writing clarity and structure. It runs locally via Ollama and catches 40% more issues than Grammarly.

x.com/May 24, 2026/3 min read
open sourcenlpacademic writing
Andrej Karpathy speaking on a YouTube video, with a diagram of neural networks and CPUs shown behind him
Opinion & Analysis
85

Karpathy: Neural nets will become the host, CPUs the co-processor

Karpathy predicts neural networks will become the host OS, with CPUs as co-processors, rendering most classical app interfaces obsolete.

x.com/May 23, 2026/3 min read
karpathyai theorycomputing paradigms
A data center rack of Cerebras wafer-scale AI chips with large cooling systems, processing the Kimi K2.6 model at…
Products & Launches
93

Cerebras Hits 981 Tokens/sec on 1T-Parameter Kimi K2.6, Claims 6.7× GPU Cloud Speedup

Cerebras reported 981 tokens/sec on the 1T-parameter Kimi K2.6 model, a 6.7× speedup over the next GPU cloud, validated by an independent third party.

x.com/May 23, 2026/3 min read/Multi-Source
cerebrasbenchmarkmoonshot ai
A person wearing a smartwatch on their wrist, with a glowing digital interface showing heart rate and activity data…
AI Research
75

Google Paper: Wearable AI Needs Personalization to Work

Google paper shows 18% heart rate accuracy gain by personalizing wearable AI to individual users via lightweight embeddings.

x.com/May 23, 2026/3 min read
personalizationwearablesgoogle
A researcher's screen displays Google Vertex AI dashboard listing Anthropic Opus 4.8 model, with a chart showing…
AI Research
91

Anthropic Opus 4.8 Spotted on Google Vertex, Sonnet 4.8 Coming Soon

Opus 4.8 spotted on Vertex AI. Unconfirmed but plausible given GPT-5.5 pressure. Watch for official confirmation.

x.com/May 23, 2026/3 min read/Multi-Source
anthropicai modelsgoogle cloud
A bar chart showing 88.5% average improvement across 126 settings, with a frozen LLM icon and code interface harness…
AI Research
84

Code-as-Agent Harness Thesis: 88.5% Gains Without Touching the LLM

Paper shows 88.5% improvement by adapting runtime interface around frozen LLM. Harness generalizes across 18 backbones, challenging model-centric agent improvement.

x.com/May 23, 2026/3 min read
llm agentsproduction aiai research