TLDRocket
Sign in
Latest Meet the 82-year-old Kentucky grandma who turned down $26 million to t... — Fortune How Chevron became the AI darling of Big Oil — Fortune Mistral AI Releases Shieldstral 1.0 3B: An Open-Weights Policy-Adaptiv... — MarkTechPost [AINews] Zawinski's Law of MultiAgents — Latent Space Now we have a timeline of the OpenAI accidental attack against Hugging... — Simon Willison’s Weblog OpenAI says it slowed Astra model development over security concerns — TechCrunch Tencent Cloud Open-Sources TencentDB Agent Memory v2.0: A Team-Level M... — MarkTechPost Auto Mode will soon be the default in Claude Code — because humans can... — The New Stack

The AI intelligence platform

Every AI story that matters and the intelligence behind it.

TLDRocket reads all relevant sources, removes duplicate coverage, and publishes a short neutral summary of every story, linking back to the original. Free, no spam, unsubscribe anytime.

Add to Slack

Every story also updates live profiles event timelines weekly rankings the AI Market Index

Tuesday, 18 November 2025

Start building with Gemini 3

Google DeepMind 8 months ago 31 3 sources

Google released Gemini 3 Pro, an AI model designed to assist developers with coding and app creation across multiple platforms. The model scores 54.2% on Terminal-Bench 2.0 for tool use and costs $2 per million input tokens through the Gemini API. Developers can now use Gemini 3 Pro in Google AI Studio, Android Studio, and third-party tools like Cursor and GitHub Copilot, with a new Google Antigravity platform enabling autonomous agents to handle complex development tasks.

We’re expanding our presence in Singapore to advance AI in the Asia-Pacific region

Google DeepMind 8 months ago 7

Google DeepMind is opening a new research lab in Singapore to expand its presence in the Asia-Pacific region. The company has more than doubled its APAC team over the past year and will focus the Singapore lab on advancing Gemini capabilities, linguistic inclusivity for Asian languages, and collaboration with regional governments and institutions. The expansion aims to tailor AI technologies to serve the diverse needs of Asia-Pacific communities and build on existing applications like using AlphaFold to study Parkinson's disease and developing multilingual AI models for Southeast Asian languages.

Three Years from GPT-3 to Gemini 3

One Useful Thing 8 months ago 21 3 sources

Google launched Gemini 3, which the author tested by having it demonstrate AI progress through building an interactive game and performing research tasks that would typically be assigned to PhD students. The model completed a 14-page research paper analyzing crowdfunding data using original statistical methods and novel hypotheses with minimal guidance. The shift from AI as chatbot to AI as autonomous agent capable of planning, coding, and executing complex work with human oversight represents the most significant change since ChatGPT's release three years ago.

A new era of intelligence with Gemini 3

Google DeepMind 8 months ago 45 3 sources

Google released Gemini 3, its latest AI model combining multimodal understanding, reasoning, and agentic capabilities across its product suite. Gemini 3 Pro scored 1501 Elo on the LMArena Leaderboard and achieved 91.9% on GPQA Diamond, while the enhanced Gemini 3 Deep Think mode reached 45.1% on ARC-AGI-2. The model is now available in Search with generative UI features, the Gemini app, developer platforms including the new Google Antigravity, and Vertex AI for enterprises.

Generative UI: A rich, custom, visual interactive user experience for any prompt

Google Research 8 months ago 32 2 sources

Google introduced generative UI, where AI models create complete interactive user interfaces and applications in response to prompts, rather than just text content. Human evaluators strongly preferred interfaces generated by Google's implementation over standard LLM outputs and top search results, though generation currently takes a minute or more. The capability is now available in Google's Gemini app through dynamic view and visual layout experiments, plus in Google Search's AI Mode for paid subscribers.

Intuit and OpenAI join forces on new AI-powered experiences

OpenAI 8 months ago 32 2 sources

Intuit and OpenAI entered a multi-year partnership valued at $100 million or more to integrate Intuit applications into ChatGPT and develop personalized financial tools using OpenAI's models. The deal includes embedding Intuit's apps directly within ChatGPT, allowing users to access financial services through the chatbot interface. Intuit gains access to advanced AI models to enhance its software offerings while OpenAI gains a major enterprise customer for its commercial services.

The daily briefing

Every AI story that matters, in your inbox by 8am.

TLDRocket reads all relevant sources, removes duplicate coverage, and summarises the day in two minutes. Follow companies and topics for alerts, or get the briefing in Slack. Free, no spam, unsubscribe anytime.