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.

GPT-5.5 Pro Leapfrogs on Epoch Benchmark; Base Model Beat…

AI Research
99

GPT-5.5 Pro Leapfrogs on Epoch Benchmark; Base Model Beats Prior Pro

A tweet from @kimmonismus reveals GPT-5.5 Pro shows significant Epoch benchmark gains, and the non-Pro GPT-5.5 surpasses GPT-5.4 Pro, suggesting major efficiency improvements at OpenAI.

x.com/Apr 29, 2026/3 min read/Multi-Source
gpt-5.5benchmarksai efficiency
Xiaomi MiMo 2.5 Pro logo beside Arena leaderboard chart, with MIT open-source badge and Opus 4.5 ranked lower
AI Research
97

Xiaomi MiMo 2.5 Pro Beats Opus 4.5 on Arena, MIT License

Xiaomi's MiMo v2.5 Pro, an open-source model under MIT license, has achieved a higher Arena score than Opus 4.5, signaling a major shift in competitive AI performance.

x.com/Apr 29, 2026/3 min read/Multi-Source
open sourcexiaomibenchmarks
Three Chinese AI company logos—Alibaba, ByteDance, Zhipu AI—alongside US and French tech logos arranged on a digital…
Big Tech
74

Time's First AI A-List: Alibaba, ByteDance, Zhipu AI Make Cut

Time magazine named Alibaba, ByteDance, and Zhipu AI among its first AI-specific top 10 list, alongside six US companies and France's Mistral AI. The recognition highlights China's growing global influence through open-source models and consumer AI apps.

scmp.com/Apr 29, 2026/3 min read
time magazinealibabachina ai
Scatter plot with regression line showing strong negative correlation between embedding distance and attack success…
AI Research
82

Embedding distance predicts VLM typographic attack success (r=-0.93)

A new study shows that embedding distance between image text and harmful prompt strongly predicts attack success rate (r=-0.71 to -0.93). The researchers introduce CWA-SSA optimization to recover readability and bypass safety alignment without model access.

arxiv.org/Apr 29, 2026/3 min read/Multi-Source
red teamingvlm safetytypographic injection
A complex neural network diagram with interconnected nodes and memory blocks, illustrating the Large Memory Model…
AI Research
87

Large Memory Models: New Architecture Beyond RAG and Vector Search

Researchers with 160+ Nature and ICLR publications have built Large Memory Models (LMMs), a new architecture designed to emulate human memory processes, offering an alternative to RAG and vector search paradigms.

x.com/Apr 29, 2026/3 min read
architecturememoryvector search
A sleek dark laptop screen displays a Mistral AI tweet announcing the Mistral Medium model, with a minimalist…
Products & Launches
86

Mistral Medium Model Launch Teased by European AI Company

Mistral AI teased an upcoming model called Mistral Medium on X, signaling continued expansion of its model lineup. The announcement comes amid growing competition in the open-weight LLM space.

x.com/Apr 29, 2026/3 min read
mistral aillmeuropean tech
OpenAI team presents GPT-5.5 on a large screen, highlighting its super app strategy, 35x cost reduction on GB200…
Products & LaunchesBreakthrough
100

GPT-5.5 Launches: The Super App Strategy, Not the Model

OpenAI released GPT-5.5, codenamed Spud, 48 days after GPT-5.4. The model itself is less interesting than the super app strategy, 35x cost reduction on GB200 hardware, and 48-day release cadence that signals a deliberate acceleration.

pub.towardsai.net/Apr 28, 2026/3 min read/Multi-Source
gpt-5.5agentic ainvidia
A researcher points at a whiteboard diagram showing three layers of JSON structure for AI skill representation, with…
AI Research
82

SSL: Structured Skill Language Boosts Skill Discovery MRR to 0.707

Researchers propose SSL, a three-layer typed JSON representation for AI agent skills, replacing unstructured SKILL.md prose. Using an LLM normalizer, SSL improves Skill Discovery MRR from 0.573 to 0.707 and Risk Assessment macro F1 from 0.744 to 0.787 on a newly released 6,184-skill corpus.

x.com/Apr 28, 2026/3 min read
structured representationdiscoveryllm
A researcher gestures excitedly while discussing with colleagues as an AI agent displays complex scientific data on…
AI Research
85

OpenAI Agents Now Ask Questions Good Enough for Research Papers

Sébastien Bubeck revealed on the OpenAI Podcast that internal AI agents now ask research questions so insightful they're inspiring papers and correcting published mistakes, with a 1-2 year timeline for full researcher-level capabilities.

x.com/Apr 28, 2026/3 min read
scientific discoveryopenaiagent systems
A Tuna-2 diagram shows raw pixels flowing into a multimodal model without a separate encoder, outperforming…
AI Research
90

Meta Tuna-2: Encoder-Free Multimodal Model Beats VAE-Based Rivals

Meta released Tuna-2, an encoder-free multimodal model that understands and generates images from raw pixels. It beats encoder-based models on fine-grained perception benchmarks, challenging the dominant VAE/vision encoder paradigm.

x.com/Apr 28, 2026/3 min read
architecturecomputer visionresearch
SandboxAQ CEO Jack Hidary stands beside a glowing digital molecular structure, representing LQMs for physics and…
Products & Launches
85

SandboxAQ Raises $950M+ for LQMs to Simulate Physics and Chemistry

SandboxAQ has raised over $950M and is backed by NVIDIA to build Large Quantitative Models (LQMs) that simulate physics and chemistry, aiming to invent new drugs and materials beyond the reach of LLMs.

x.com/Apr 28, 2026/3 min read
materials sciencefundingai
Two men, likely Alec Radford and David Duvenaud, stand beside a glowing holographic display of a historical figure…
AI Research
75

Alec Radford's 'Talk to the Past' AI Lets You Chat with History

A new AI project by Alec Radford and David Duvenaud lets you chat with simulated historical figures.

x.com/Apr 28, 2026/3 min read
researchaiconversational ai
NVIDIA logo and Nemotron 3 Nano Omni graphic highlighting unified video, audio, image, and text processing
AI Research
93

NVIDIA Nemotron 3 Nano Omni: Open Multimodal Model Unifies Video, Audio, Image, Text

NVIDIA announced Nemotron 3 Nano Omni, an open multimodal model that processes video, audio, images, and text in a unified architecture, expanding accessibility for multimodal AI research.

x.com/Apr 28, 2026/3 min read/Multi-Source
open sourceedge computingnvidia
A swarm of small, glowing AI agents surrounds a larger central model, symbolizing adversarial training to distill a…
AI Research
86

Vibe Training: SLM Replaces LLM-as-a-Judge, 8x Faster, 50% Fewer Errors

Plurai introduces 'vibe training,' using adversarial agent swarms to distill a small language model (SLM) for evaluating and guarding production AI agents. The SLM outperforms standard LLM-as-a-judge setups with ~8x faster inference and ~50% fewer evaluation errors.

x.com/Apr 28, 2026/3 min read
model distillationslmplurai
Data center racks with glowing server lights, suggesting high CPU density for AI workloads
AI Research
96

JPMorgan: Agentic AI Could Flip Server Ratio to CPU-Heavy

JPMorgan reports that agentic AI workloads could increase CPU demand, potentially flipping the GPU-to-CPU ratio from 7-8 GPUs per CPU to CPU-heavy deployments, with a $100B TAM for AI CPU infrastructure.

x.com/Apr 28, 2026/3 min read
ai infrastructuredata centershardware
Chinese employees in an office using OpenClaw AI agent on screens, with Shenzhen subsidy documents visible, while a…
Products & Launches
77

China's OpenClaw Mandate: Subsidies, Quotas, and Firing for Non-Use

In China, OpenClaw ('raising lobsters') is subsidized by Shenzhen and mandated for daily employee tasks, with non-use leading to termination. Meanwhile, using OpenAIClaw elsewhere risks firing. This signals a stark AI adoption divide.

x.com/Apr 28, 2026/3 min read
open sourcechinaai agents
A computer screen shows a GitHub repository page for The Agency, displaying 147 AI agents across 12 divisions with…
Products & Launches
86

The Agency: 147 Open Source AI Agents Hit 50K GitHub Stars in 2 Weeks

The Agency is an open source repository with 147 specialized AI agents across 12 divisions (engineering, design, marketing, etc.) that hit 50K GitHub stars in under two weeks. It provides one-command install for tools like Claude Code and Cursor, with full modding support.

x.com/Apr 28, 2026/3 min read
open sourcegithubai agents
Three interconnected memory systems labeled parametric, retrieval, and agent memory displayed on a diagram with a…
AI Research
80

AI Memory Survey: Three Systems Needed for Human-Like Recall

A new survey paper proposes that modern AI requires three distinct memory systems—parametric, retrieval, and agent memory—to achieve human-like cognition, highlighting control as the key bottleneck.

x.com/Apr 28, 2026/3 min read
survey paperagentsmemory systems
Google executives shaking hands with Pentagon officials in a government meeting room, with a laptop showing AI data…
Products & Launches
95

Google Inks Pentagon AI Deal, Reverses 2018 Stance

Google signed a deal allowing the Pentagon to use its AI models for classified work and 'any lawful government purpose,' reversing its 2018 exit from Project Maven. The contract includes non-binding language on surveillance and autonomous weapons, and requires Google to adjust AI safety filters at government request.

x.com/Apr 28, 2026/3 min read
ai ethicsclassified contractsmilitary ai
A data privacy dashboard with highlighted PII categories like names, emails, and medical IDs, overlaid with a shield…
AI Research
85

OpenAI Privacy Filter Gets 6x More PII Labels via Nvidia Data

OpenAI has retrained its privacy filter using Nvidia's Nemotron-PII dataset, expanding PII detection from 8 to over 50 label types, targeting healthcare and enterprise use cases with better accuracy.

x.com/Apr 28, 2026/3 min read
privacynvidiaopenai
A 3D-rendered video scene shows a realistic physics simulation of a ball rolling down a staircase with accurate…
AI Research
85

Microsoft World-R1: RL Aligns Text-to-Video with 3D Physics

Microsoft's World-R1 framework applies reinforcement learning with feedback from pre-trained 3D foundation models to align text-to-video outputs with physical 3D constraints, improving structural coherence without modifying the underlying video diffusion architecture.

x.com/Apr 28, 2026/3 min read
3d aiopen-sourcereinforcement learning
OpenAI logo on a dark background with AWS, Google Cloud, and Microsoft Azure cloud icons arranged around it…
Products & Launches
89

OpenAI Breaks Microsoft Exclusivity, Eyes AWS and GCP

OpenAI is moving away from its exclusive Microsoft cloud arrangement, signaling potential partnerships with Amazon AWS and Google Cloud to diversify infrastructure and reduce dependency.

x.com/Apr 28, 2026/3 min read
ai infrastructuremicrosoftcloud computing
A diagram comparing standard language model output with KARL's reinforcement learning approach, showing a knowledge…
AI Research
76

KARL: RL Framework Cuts LLM Hallucinations Without Accuracy Loss

KARL introduces a reinforcement learning framework that dynamically estimates an LLM's knowledge boundary to reward abstention only when appropriate, achieving a superior accuracy-hallucination trade-off on multiple benchmarks without sacrificing correctness.

arxiv.org/Apr 28, 2026/3 min read
ai safetyresearchllm
Microsoft logo and Playwright code on a screen, an AI agent interacts with a web page via accessibility tree data…
Products & Launches
100

Microsoft's Playwright MCP Server Replaces Vision for Web Agents

Microsoft built an MCP server for Playwright that lets AI agents interact with web pages using the accessibility tree, eliminating the need for screenshots and vision models. This approach reduces hallucinations and broken selectors, working with tools like Cursor, VS Code, and Claude Desktop.

x.com/Apr 28, 2026/3 min read/Widely Reported
browser automationmcpai agents
A vintage-style computer monitor displays glowing green text from an old language model, surrounded by antique books…
AI Research
85

Talkie: Vintage LLM Trained on 260B Pre-1931 English Tokens

Talkie is a new 'vintage language model' trained on 260 billion tokens of historical English text from before 1931, developed by a team including Alec Radford, co-author of the original GPT paper. It offers a unique linguistic artifact for NLP research.

x.com/Apr 28, 2026/3 min read
alec radfordhistorical aivintage ai
A crowd of tech workers in a corporate lobby, some holding boxes and looking at a digital screen displaying AI icons…
Products & Launches
85

78,557 Tech Workers Laid Off in Q1 2026; Nearly Half Replaced by AI

A new paper reports 78,557 tech layoffs in Q1 2026, with nearly half of those roles replaced by AI automation, marking a significant shift in workforce dynamics.

x.com/Apr 28, 2026/3 min read
tech industryworkforcelayoffs

Microsoft’s VibeVoice: Open-Source Speech-to-Text with Di…

Products & Launches
85

Microsoft’s VibeVoice: Open-Source Speech-to-Text with Diarization

Microsoft released VibeVoice, an MIT-licensed speech-to-text model with built-in speaker diarization. Simon Willison tested a 4-bit MLX conversion on an M5 MacBook, transcribing 1 hour of audio in ~9 minutes using ~60GB RAM.

x.com/Apr 27, 2026/3 min read
open sourcemachine learningai
Smartphone screen showing Gemini app interface with a new file directory folder icon, a document creation button…
Products & Launches
77

Gemini App Gets File Creation and Its Own File Directory

The Gemini app now supports file creation and a dedicated file directory, enabling users to work directly within the app. This transforms Gemini from a conversational AI into a more autonomous workspace tool.

x.com/Apr 27, 2026/3 min read
productivity toolsfile managementgoogle gemini
OpenAI and Microsoft logos on a smartphone screen displaying ChatGPT, symbolizing their updated partnership ahead of…
Products & Launches
91

OpenAI Drops AGI Clause with Microsoft Ahead of IPO

OpenAI has removed the AGI clause from its Microsoft partnership, ending restrictions that limited Microsoft's access to future AGI systems. The move, reported ahead of OpenAI's anticipated IPO, suggests OpenAI may be preparing to announce AGI milestones.

x.com/Apr 27, 2026/3 min read
ai businessmicrosoftopenai
Sleek website interface for MiniMax M2 AI music generator with text input field and waveform visualization on…
Products & Launches
75

MiniMax Music-2.6 Goes Free on Cloudflare This Week

MiniMax's Music-2.6 AI model is available for free on Cloudflare's platform this week, allowing users to generate full-length songs or instrumentals from text prompts.

x.com/Apr 27, 2026/3 min read
ai musicminimaxai models