Do Son Google proposes the ai-disclosure tag in Jan 2026. Discover how this new Chrome standard will label AI-generated text across the web.
machine-learning
Google launches Personal Intelligence in Jan 2026! Search AI Mode now uses your Gmail and Photos to provide hyper-tailored results.
Wikipedia marks 25 years on Jan 15, 2026, by announcing major AI data deals with Microsoft, Meta, and Amazon to fund its non-profit human knowledge mission.
Google updates Gemini limits in Jan 2026, tripling Thinking model prompts to 300 for Pro users and launching a 1,500-prompt Ultra tier for $250/mo.
Bixby is back: Samsung’s One UI 8.5 beta reveals a deep Perplexity AI integration, bringing context-aware reasoning to the Galaxy S26 and beyond.
MIT and NVIDIA unveil FoundationMotion, an automated system that enables AI to master complex video motion, outperforming Gemini in autonomous driving scenes.
Google is rebranding the Pixel 10's Magic Cue as Contextual Suggestions, bringing proactive, privacy-first AI to all Android devices via Play Services.
Google’s Gemini 3 Flash defies logic: an ultra-efficient model scoring 81.2% on MMMU-Pro, outperforming OpenAI’s flagship GPT-5.2 in multimodal reasoning.
Critical Ray AI flaw (CVE-2025-62593) allows RCE via DNS rebinding on Safari & Firefox. Chrome is immune. Update to Ray v2.52.0 immediately.
Canonical and NVIDIA are collaborating to bring the CUDA Toolkit to Ubuntu's repositories, streamlining the installation process for developers.
A critical flaw (CVE-2025-6507) in the H2O-3 machine learning platform allows attackers to bypass security filters, leading to remote code execution.
xAI has launched grok-code-fast-1, a new AI coding model optimized for speed and efficiency. Learn how it plans to challenge Microsoft and OpenAI.
YouTube has admitted to secretly using AI to "enhance" Shorts videos without creators' consent, a move that has sparked backlash over transparency and trust.
OpenAI has released GPT-5, touting major advances in reasoning, speed, and coding. The new model is now available to developers and enterprise customers.
OpenAI's GPT-5 is officially launching! Get the details on the different versions, from GPT-5 Mini to GPT-5 Chat, and how they will be rolled out.
A critical SSRF vulnerability (CVE-2025-54381) in BentoML's file upload handling allows unauthenticated remote attackers to perform internal network reconnaissance and steal cloud metadata credentials from AI applications.
ChatGPT agents are now capable of bypassing Cloudflare Turnstile CAPTCHA verification, automatically clicking checkboxes and demonstrating a new level of AI-driven anti-bot system evasion.
GPUHammer is the first Rowhammer attack on GDDR6 memory (NVIDIA A6000), successfully inducing bit flips that can degrade AI model accuracy. Enable System-Level ECC to mitigate.
MIT researchers unveiled an AI-powered technique for reversible restoration of historical paintings, creating a "damage map" and reproducing colors on a removable film.
A critical CVSS 9.8 flaw (CVE-2025-49653) in Backend.AI allows account takeover via PoC, impacting all versions. Vendor has not issued a patch.
Google Cloud launches Cloud Run GPU, offering auto-scaling NVIDIA L4 GPUs for AI workloads with per-second billing and rapid cold start.
Google AI Edge Gallery brings powerful, privacy-focused AI models to Android devices (iOS coming soon!). Run AI offline, protect your data, and explore features like Prompt Lab.