AI News — Sunday, September 20, 2026

10
Google's Gemini AI Model Demonstrates Hacking Capabilities Against Other Companies

Google's Gemini AI model has reportedly been used to hack other companies, raising significant concerns about AI security and ethical deployment.

TechCrunchindustry
9
MiniMax-H3 Evaluates Omni-Modal Generative Model's Physical World Reasoning

A new paper evaluates MiniMax-H3, an omni-modal generative model, on its ability to reason about the physical world, showing advancements in AI's understanding of real-world physics.

Hugging Faceresearch
9
Trump Proposes AI Rebranding and Creation of an 'AI Force'

Former President Trump suggests rebranding AI with a new name and announces plans to establish an 'AI Force,' indicating a potential shift in national AI strategy and terminology.

TechCrunchindustry
8
JEPA-Anything: Learning Predictive Models Across Diverse Environments

Researchers introduce JEPA-Anything, a novel approach for learning predictive models that can generalize and adapt across various simulated or real-world environments.

Hugging Faceresearch
8
AI in Interviews: Applicant Rejected for AI Use, Interviewer Uses It

A developer recounts being rejected from an interview for using AI, only to observe the interviewer subsequently using AI themselves, highlighting a double standard in the industry.

Dev.toindustry
8
AI Is Finding Vulnerabilities Faster Than Humans Can Patch Them – A Security Crisis

The rapid pace at which AI discovers software vulnerabilities is outpacing human patching capabilities, creating a growing security crisis across industries.

Dev.toindustry
7
Verifiable Social Reasoning for LLM Assistants

A new study explores methods for enabling LLM assistants to perform verifiable social reasoning, enhancing their reliability and trustworthiness in complex human interactions.

Hugging Faceresearch
7
Flock Reportedly Attempts Workforce Reduction Through Employee Buyouts

AI startup Flock is reportedly offering employee buyouts in an effort to shrink its workforce, signaling potential challenges or strategic shifts within the company.

TechCrunchindustry
7
Co-creating the Future of Fashion with Google AI

Google announces collaborations to integrate AI into the fashion industry, aiming to co-create future trends and design processes.

Google AI Blogproduct
7
I Let AI Write My Tests for 6 Months: What Survived Production

A developer shares insights from six months of using AI to write tests, detailing which AI-generated tests proved robust enough for production environments.

Dev.toindustry
6
RiskChainBench: A Benchmark for Obfuscated Message Restoration and Web Investigation

Researchers introduce RiskChainBench, a new benchmark designed to evaluate AI models' ability to restore obfuscated platform messages and conduct evidence-grounded web investigations.

Hugging Faceresearch
6
Self-Evolving Search Index Proposed for Dynamic Information Retrieval

A new paper presents the concept of a self-evolving search index, which dynamically adapts and improves its structure to enhance information retrieval efficiency over time.

Hugging Faceresearch
6
Reflect, Revise, Reuse: Training-Free Skill Evolution for GUI Agents

A new research paper introduces a training-free method for GUI agents to evolve their skills through a process of reflection, revision, and reuse, improving adaptability.

Hugging Faceresearch
6
PACT: Can Enterprise AI Assistants Be Trusted Under Pressure?

Research investigates the trustworthiness of enterprise AI assistants when operating under high-pressure scenarios, assessing their reliability and decision-making capabilities.

Hugging Faceresearch
5
Video DeltaNet: A Video-Native Hybrid Attention for Livestream Video Generation

Video DeltaNet is proposed as a novel video-native hybrid attention model designed to improve the quality and efficiency of livestream video generation.

Hugging Faceresearch