Perplexity Now Available as a Search Option in Firefox 144

Perplexity has expanded its user access through a new integration with the Mozilla Firefox browser.

NanoChat: A Clear and Accessible Blueprint for Building ChatGPT-Style Models

The best ChatGPT that $100 can buy

Elon Musk’s xAI Joins Race to Build ‘World Models’ to Power Video Games

Elon Musk's AI venture recruits Nvidia experts to advance systems capable of understanding and creating real-world environments

Microsoft Rolls Out Copilot Update for Windows Insiders: Connectors and Document Creation Features

Microsoft has started rolling out an update to the Copilot app on Windows for Insiders

Figure 03: A Step Forward in Humanoid Robotics

Figure 03, is advancing what’s possible. Designed for homes, commercial spaces, and large-scale production, Figure 03 brings practical robotics closer to reality.

New Open-Source Robotics Repository from MathWorks Offers Hands-On Learning for All

MathWorks has launched an exciting open-source GitHub repository filled with robotics resources, making it easier than ever for enthusiasts, students, and researchers to dive into building and experimenting with robots.

Beyond the Lab: DEEP Robotics Introduce A Humanoid Robot for Real-World Challenges.

DEEP Robotics has officially launched the DR02, touting it as the world's first industrial-grade humanoid robot.

Google Unveils Gemini 2.5 Computer Use: A Pro Model Built for UI Agent Control

Google has introduced Gemini 2.5 Computer Use, a refined version of its Gemini 2.5 Pro model designed specifically for agents that interact directly with user interfaces on the web and mobile platforms.

xAI Rolls Out Imagine v0.9: Key Upgrades in Video and Audio Generation

xAI has released Imagine v0.9, a substantial update to its video generation model that builds on the foundation of v0.1.

OpenAI Introduces Apps in ChatGPT and a New Apps SDK

OpenAI has announced a major update to ChatGPT, the introduction of chat-based apps and a new Apps SDK that allows developers to build them.

Landing AI Introduces DPT-2: A Smarter Way to Handle Messy Documents

Document extraction is one of those practical AI challenges that rarely gets the spotlight but makes a huge difference in real-world automation.

OpenAI Refines Sora with Enhanced Creator Controls and Revenue Sharing

OpenAI's video generation tool, the company outlined two immediate adjustments aimed at addressing real-world usage patterns and stakeholder input.

Anthropic Launches Claude Integration with Slack to Streamline Team Workflows

Anthropic has announced new integrations between its Claude assistant and Slack

DoorDash Rolls Out Autonomous Delivery Robot Dot

DoorDash has launched its own autonomous delivery robot, named Dot, marking a step into robotics for the food delivery giant

Thinking Machine Announced Tinker: An API for open-model training.

Thinking Machines Lab, the team behind the historical TinkerToy computer, is now entering this practical arena with a new service named Tinker.

Microsoft Launches Vibe Working: Copilot's Agents – The End of Tedious Task Tango

Microsoft announcement introduces \"Agent Mode\" and \"Office Agent\" to Microsoft 365 Copilot.

xAI's Grok Coding Models Dominate OpenRouter Leaderboard

xAI's Grok coding models are making waves, topping the OpenRouter leaderboard

OpenAI Unveils Sora 2: Bringing More Realistic Video Creation to Life

OpenAI has just released Sora 2, their updated video and audio generation model.

Anthropic Unveils Claude Sonnet 4.5: AI Coding and Reasoning King

Anthropic announced the release of Claude Sonnet 4.5, positioning it as their most advanced model yet for coding, building complex AI agents, and computer interaction.

The Wuji Hand: A Robotic Hand Developed by Pan Motor

Pan Motor, a company based in China, has developed the Wuji Hand, a human-sized robotic hand

Discovering Forpheus: The AI Table Tennis Robot Revolutionizing the Game

Perplexity Addresses Programmatic Access to Web Information with New Search API

Perplexity has now introduced its Search API to tackle this issue. The API grants developers access to the same infrastructure that supports Perplexity's public answer engine

Moonshot AI Launches OK Computer Mode for Kimi Assistant

Kimi, a prominent AI platform, recently introduced “OK Computer”, a new agent mode designed to act more like a collaborative team member than a chatbot.

ChatGPT now thinks ahead by introducing Pulse.

OpenAI has launched Pulse, a new feature in ChatGPT that is now available to Pro subscribers

Google Labs Introduces Mixboard: An AI-Powered Tool for Idea Visualization

Google Labs has launched Mixboard, an experimental platform described as a new way to visualize ideas.

Claude Bridges Design and Code with Figma Integration

Anthropic has announced a new capability that connects its Claude code system with Figma through the Model Context Protocol (MCP).

OpenAI Hits Perfect Score at ICPC 2025, Following Gemini’s Historic Gold-Level Run

OpenAI did excellently well at the IPCC 2025 competition, this exceptional performance is huge and remarkable.

Grok 4 Fast: Smaller, Smarter, Cheaper A New Standard in Efficient AI

Grok 4 Fast builds on the success of Grok 4 but focuses on something critical for today’s developers and enterprise users which is cost-efficiency without compromise.

NCAIR Advances Indigenous Language AI with Open-Source Speech Models

The Nigeria’s dedicated hub for AI and 4IR technologies under NITDA, has taken another important step in expanding access to inclusive digital tools

Gemini 2.5 Deep Think Achieves Gold-Medal Level at ICPC World Finals

Gemini 2.5 Deep Think has reached gold-medal level performance at the 2025 International Collegiate Programming Contest (ICPC) World Finals

Anthropic Shares Postmortem on Infrastructure Bugs That Impacted Claude

Anthropic has published a detailed postmortem analyzing three infrastructure issues that recently degraded the quality of responses from its Claude models.

Google unveils Agent Payments Protocol (AP2) for secure agent-led commerce

Google introduced the Agent Payments Protocol (AP2), an open, payment-agnostic framework that lets AI agents securely initiate and complete transactions across platforms. Built to work alongside the Agent2Agent (A2A) protocol and Model Context Protocol (MCP), AP2 defines how authorization, authenticity, and accountability are handled when an agent triggers a purchase.

World Labs Unveils Marble: A New Step Toward Persistent 3D Worlds

World Labs has announced Marble, a limited access beta preview of its latest model for generating persistent, navigable, and controllable 3D worlds.

Claude Now Available in Xcode 26 for Developers

Developers can now connect their Claude accounts directly to Xcode 26

Largest Study to Date Shows How People Are Using ChatGPT

OpenAI’s Economic Research team, together with Harvard economist David Deming, has released the largest study so far on ChatGPT consumer usage. P

VaultGemma: A New Benchmark for Privacy-Focused AI

Gemma has introduced VaultGemma, described as the largest open large language model (LLM) trained from scratch with differential privacy (DP).

Manus AI Unveils Connectors to Bridge Tools and Eliminate Silos

Manus AI has introduced Manus Connectors, a new feature designed to integrate its AI capabilities with users' existing tools, reducing the friction of context switching across multiple applications

Ground Control Robotics Advances Multi-Legged Robot Locomotion in Collaboration with Georgia Tech

Ground Control Robotics is partnering with the Georgia Institute of Technology to develop a new locomotion paradigm tailored for cluttered and confined spaces.

Anthropic Adds Memory Feature to Claude for Workplace Teams

Anthropic has introduced a memory feature to its Claude app, designed to help users and teams maintain context across conversations.

Introducing Agent 3: A New Era of Autonomy in App Building

Replit has unveiled Agent 3, the latest version of its AI-powered Agent, designed to give developers more time back while ensuring smoother, more reliable app creation.

South Korea Establishes National AI Strategy Committee

South Korea took a step forward in its AI efforts by launching the National AI Strategy Committee.

Qwen3-ASR-Flash: Advancing Speech Recognition Across Languages and Contexts

Alibaba’s Qwen team has introduced Qwen3-ASR-Flash, a new automatic speech recognition (ASR) service built on the intelligence of Qwen3-Omni and trained on tens of millions of hours of multimodal data.

Claude AI Enhances Mobile App with Real-World Integration Features

Anthropic's Claude AI has introduced new capabilities that allow users to connect the AI more directly with their physical and digital environments.

Researchers Demonstrate Rapid Adaptation to Dani Clode’s Third Thumb

At the Royal Society’s Summer Science Exhibition, researchers from the MRC Cognition and Brain Sciences Unit at the University of Cambridge

Italian Restaurant Sends Pizza by Drone in Potential Glimpse of Delivery's Future

A restaurant in Italy has reportedly delivered a hot pizza via drone, flying it across town directly to a customer no car, no driver, just the hum of cutting-edge tech weaving above rooftops and streets.

DeepRobotics Signals Scale in Quadruped Robotics

DeepRobotics has announced that its quadruped robots are now in mass production and deployed globally, marking a significant milestone for the company and for enterprise automation more broadly.

NotebookLM Expands Audio Overview Options

NotebookLM has introduced 4 new audio overview formats, giving users more flexible ways to engage with their source materials.

Projects in ChatGPT: Organize and Continue Your Work Seamlessly

OpenAI has introduced Projects in ChatGPT, a feature designed to help users keep their long-running work organized in one place.

Google Introduces Custom Slash Commands in Gemini CLI

Google has added support for custom slash commands in its Gemini CLI tool.