AI News — Friday, May 22, 2026

9
Spotify and Universal Music Strike Deal Allowing Fan-Made AI Covers and Remixes

Spotify and Universal Music Group have partnered to permit fan-created AI covers and remixes, signaling a significant shift in music rights and AI integration.

TechCrunchindustry
9
I/O 2026: Welcome to the Agentic Gemini Era

Google's I/O 2026 keynote introduces the 'agentic Gemini era,' highlighting a future where AI agents powered by Gemini play a more autonomous and integrated role.

Google AI Blogindustry
8
Trump Delays AI Security Executive Order, Citing Potential 'Blocker' Language

Former President Trump has postponed an AI security executive order, citing concerns that its initial wording might hinder innovation.

TechCrunchindustry
8
Salesforce Rolls Out New Slackbot AI Agent to Compete in Workplace AI Market

Salesforce has launched a new Slackbot AI agent, intensifying its competition with Microsoft and Google in the enterprise workplace AI market.

VentureBeatproduct
8
Nous Research's NousCoder-14B: An Open-Source Coding Model for the 'Claude Code Moment'

Nous Research has released NousCoder-14B, an open-source coding model positioned as a strong competitor in the rapidly evolving AI code generation space.

VentureBeatopen-source
8
Mega-ASR: Scaling Real-world Acoustic Simulation for In-the-wild Speech Recognition

Researchers introduce Mega-ASR, a new approach to significantly improve speech recognition in diverse real-world conditions by scaling up acoustic simulation.

Hugging Faceresearch
7
AdventHealth Advances Whole-Person Care with OpenAI Partnership

AdventHealth is partnering with OpenAI to integrate AI into its healthcare services, aiming to enhance whole-person care delivery.

OpenAI Blogindustry
7
Video2GUI: Synthesizing Large-Scale Interaction Trajectories for Generalized GUI Agent Pretraining

A new method, Video2GUI, enables the synthesis of vast interaction data to pretrain AI agents for generalized graphical user interface control.

Hugging Faceresearch
7
Enhancing Train-Free Infinite-Frame Generation for Consistent Long Videos

A novel technique allows for the generation of consistent, infinitely long videos without requiring additional training, advancing video synthesis capabilities.

Hugging Faceresearch
7
You Only Need Minimal RLVR Training: Extrapolating LLMs via Rank-1 Trajectories

This paper suggests that LLMs can be effectively extrapolated with minimal Reinforcement Learning from Variational Rewards (RLVR) training using rank-1 trajectories.

Hugging Faceresearch
7
OScaR: Occam's Razor for Extreme KV Cache Quantization in LLMs and Beyond

OScaR introduces an extreme KV cache quantization method for LLMs, significantly reducing memory footprint while maintaining performance.

Hugging Faceresearch
6
Introducing OpenAI for Singapore: Fostering AI Innovation in the Region

OpenAI is expanding its presence and services into Singapore, marking a strategic move to foster AI innovation and adoption in the region.

OpenAI Blogindustry
6
Building a Database Performance Testing Tool With AI: The Honest Breakdown

An article details the practical challenges and insights gained from developing an AI-powered tool for database performance testing.

Dev.toproduct
6
IndusAgent: Reinforcing Open-Vocabulary Industrial Anomaly Detection with Agentic Tools

IndusAgent proposes using agentic tools to enhance open-vocabulary industrial anomaly detection, improving robustness and adaptability in manufacturing.

Hugging Faceresearch
6
A Survey of Large Audio Language Models: Generalization, Trustworthiness, and Outlook

A comprehensive survey explores the current state, challenges, and future directions of Large Audio Language Models, focusing on generalization and trustworthiness.

Hugging Faceresearch