AI News — Sunday, November 22, 2026

10
AI Supply Chain Privacy Disaster: Researcher Plants 2M Canaries, Exposing Secret User Data

A researcher uncovered a significant privacy vulnerability in the AI supply chain by embedding 2 million 'canaries' that revealed widespread exposure of secret user data.

Lobste.rsindustry
9
Signal Restricts Mobile Screenshots Amid 'Existential Threat' of AI Surveillance

Signal has implemented restrictions on mobile screenshots, citing the growing 'existential threat' posed by advanced AI surveillance technologies.

TechCrunchindustry
8
OpenAI's GPT-6 Astra Drives Improvements in Financial Research for Hebbia

Hebbia, a financial research firm, reports significant advancements in its operations by leveraging OpenAI's latest large language model, GPT-6 Astra.

OpenAI Blogproduct
8
Can AI Help Scientists Identify Promising Drug Candidates? OpenAI Introduces MedChemBench

OpenAI introduces MedChemBench, a new initiative exploring how AI can accelerate the identification of promising drug candidates in scientific research.

OpenAI Blogresearch
7
A Tale of Two Nvidias: Analyzing the Divergent Paths of the AI Giant

An analysis delves into the contrasting aspects and strategic directions shaping Nvidia's current position and future in the AI industry.

TechCrunchindustry
7
Combatting Adversarial Distillation: OpenAI's Latest Research on AI Safety

OpenAI shares new research on 'combatting adversarial distillation,' focusing on methods to enhance the robustness and safety of AI models against malicious attacks.

OpenAI Blogresearch
7
Google Labs Introduces Tabletop: Customizable, Collaborative Spaces for Groups

Google Labs unveils Tabletop, a new platform offering customizable and collaborative digital spaces designed for groups to plan, play, and work together more effectively.

Google AI Blogproduct
6
Here Are the Top AI Agents That Can Live in Your Text Messages

A roundup highlights the leading AI agents currently available that can integrate directly into text messaging platforms for various user interactions.

TechCrunchproduct
6
HiUni: Harmonizing Understanding and Generation in Shared Visual Space

New research introduces HiUni, a model designed to improve multimodal AI by harmonizing both understanding and generation capabilities within a shared visual space.

Hugging Faceresearch
6
EgoCaptioner: Towards Comprehensive and Fine-Grained Egocentric Video Captioning

EgoCaptioner presents a novel approach aimed at generating more comprehensive and finely detailed captions for egocentric video content.

Hugging Faceresearch
6
AltRA-Test: An Alternative Annotator Test for LLM Judges Without Shared Annotators

AltRA-Test proposes a new method for evaluating LLM judges that does not require shared annotators, offering a potentially more efficient and independent assessment.

Hugging Faceresearch
5
A VS Code Extension to Turn ANY Free Chatbot Website into an AI Agent

A new VS Code extension allows users to convert any free chatbot website into a functional AI agent, expanding the utility of existing online conversational tools.

Dev.toopen-source
5
Energy-Based Reasoners: Your Recurrent Reasoner Should Be an Energy-Based Model

This research advocates for the adoption of energy-based models as the foundation for recurrent reasoners, suggesting a new paradigm for AI reasoning architectures.

Lobste.rsresearch
5
TuringSpace: Grounding Spatial Intelligence in Action with Interactive CAPTCHAs

TuringSpace explores grounding spatial intelligence in AI through interactive CAPTCHAs, aiming to improve models' understanding of physical space and action.

Hugging Faceresearch
5
ZeroVLN: Learning to Navigate without Supervised Fine-Tuning of the VLM Brain

ZeroVLN introduces a method for vision-language models to learn navigation tasks effectively without the need for extensive supervised fine-tuning.

Hugging Faceresearch