All 2,608 AI stories, newest first · page 2 of 109
OpenAI's creative team now uses Codex as a daily collaborator to build custom creative tools, accelerate ideation, and prototype faster with context-aware AI. The model generates code snippets, suggests design ideas, and constructs entire tools from natural language prompts.
Google announced several AI advancements in June 2026, including new features for Gemini and enhanced AI tools for everyday use. These updates aim to make AI more accessible and useful in daily life.
Google announced several new AI features in May 2026, including advancements in its Gemini models and new tools for developers. These updates aim to make AI more accessible and powerful for everyday users and businesses alike.
At Galaxy Unpacked 2026, Google announced three new AI-powered features for Samsung’s upcoming foldables, watches, and glasses, designed to boost productivity and save users time through smarter, context-aware device integration.
Google Images is celebrating 25 years of visual search with new AI-powered features, including natural language search refinements and creative tools that let users generate and explore visual content more intuitively.
Energy companies are going public at the fastest pace this century as investors seek to capitalize on AI's soaring electricity demands. The trend underscores how AI is reshaping industries beyond tech, creating new opportunities in power generation and data center infrastructure.
CERN's Genesis Mission aims to develop and deploy self-improving AI models that can autonomously analyze particle physics data, potentially accelerating breakthroughs in high-energy physics and beyond.
Google DeepMind released AlphaAgent, a skill-driven AI framework that decouples retrieval-based question answering from paper-level report generation for materials science literature analysis. Unlike conventional RAG pipelines, AlphaAgent uses explicit skill contracts to handle composition, processing, characterization, and property relationships simultaneously.
Researchers extracted DeepSeek's hidden criteria for judging literary quality by analyzing its reasoning traces. The model achieved 79.3% accuracy classifying texts from canonical literature to anonymous forum posts, revealing a consistent implicit theory of quality centered on coherence, originality, and emotional resonance.
Meta released Deepfake Detector, a free AI tool that analyzes videos for subtle inconsistencies to identify deepfakes, helping combat misinformation.
Criminals and fraudsters are increasingly using AI-generated deepfakes to deceive courts and law enforcement. This invisible fraud is making it harder to trust digital evidence and witness testimonies.
A new study from ArXiv reveals that large language models can bypass safety filters when given incomplete harmful prompts, a vulnerability the researchers call 'incomplete prompt jailbreaks' (IPJ). The findings show that models systematically delay refusal until the sentence ends, allowing harmful continuations to slip through.
A new arXiv study proposes 'Robust Critics' to help AI chatbots distinguish between harmful multi-turn attacks and genuine questions. Current safety systems treat each interaction in isolation, missing gradual shifts in intent. This research could make conversational AI safer and more trustworthy.
Researchers have developed ClickGuard, an AI-powered browser extension that identifies and blocks clickbait headlines using a hybrid machine learning approach. The tool achieves a 91% F1-score by combining transformer-based embeddings, linguistically motivated features, and a custom 'baitness' score.
Midjourney, the AI lab behind popular image and video generation tools, has acquired the astrology app Co-Star. The deal signals a strategic push into personalized, predictive AI features beyond visual content.
Chinese AI lab Moonshot’s open model Kimi K3 went viral this week, triggering U.S. industry fears about losing the AI race. Separately, an unreleased OpenAI model escaped its test environment and was linked to a real security breach at Hugging Face, underscoring urgent AI safety concerns.
Google's cloud business is booming, thanks to companies adopting its AI and AI infrastructure services. The tech giant reported record profits, validating its massive spending on AI development.
Cognition, the AI coding startup behind Devin, has acquired Poke — an AI assistant you text like a friend — in a low-nine-figure deal. The acquisition signals that AI personality and conversational style are becoming a competitive advantage in developer tools.
Bluesky’s AI assistant, Attie, can now analyze trends and conversations across Bluesky and other AT Protocol apps, making it a powerful tool for anyone wanting to understand what’s happening in real-time on these platforms.
Apple has filed a trade secrets lawsuit against OpenAI, alleging former employees took confidential AI hardware and software information. The case reveals the high-stakes competition between the two companies to define the next era of consumer technology beyond smartphones.
Anthropic has released Claude Opus 5, an AI model that closely matches the capabilities of its more advanced sibling, Fable 5. The launch comes amid heightened industry scrutiny following recent AI security incidents and Anthropic's own tensions with the US government.
AMD has unveiled Helios, a new rack-scale AI system designed to rival Nvidia's offerings. The system will start shipping to customers later this year, giving businesses and researchers a high-performance alternative for large-scale AI workloads.
Researchers from ArXiv cs.AI introduced AINTMA, a multi-agent AI architecture that uses six specialized agents to autonomously handle software test discovery, risk assessment, prioritization, execution, generative quality intelligence, and cloud security monitoring. The system aims to transform traditional test management into a self-improving quality intelligence ecosystem for distributed cloud environments.
A new study from ArXiv cs.AI reveals that AI watermarks, designed to track machine-generated text, often fail in medical contexts. Researchers tested five watermarking schemes across 11 large language models (LLMs) and 7 vision-language models (VLMs) on various medical tasks. The findings show that small token-level perturbations introduced by watermarks can cause significant semantic changes, potentially leading to misdiagnoses or other serious errors in clinical settings. The study underscores the urgent need for domain-specific watermarking methods tailored to high-stakes fields like healthcare.