Google's Path to AGI

Sign Up | Advertise

The AI World Today

• Google Advances Path to AGI

• Apple Integrates ChatGPT in iOS

• Deloitte Explores AI's Future

  +

• AI Training

• Heads Up

• AI Solution

Gemini 2.0 Advances Path to AGI

Google has launched Gemini 2.0, a major AI upgrade aimed at advancing generative AI from answering questions to completing tasks autonomously, according to Google DeepMind CEO Demis Hassabis. Introduced alongside the faster Gemini 2.0 Flash, the model is twice as fast as its predecessor and powers AI Overviews in Google Search, Project Astra, gaming and coding agents, and Project Mariner, a Chrome plugin for task execution. Hassabis emphasized the importance of semi-autonomous AI agents as a step toward artificial general intelligence (AGI), enabling systems to reason, plan, and act independently. While acknowledging diminishing returns in scaling large language models, Hassabis said breakthroughs in areas like reinforcement learning remain essential to achieve AGI, marking this as a significant milestone.

iOS 18.2 Features ChatGPT and Genmoji

OpenAI and Apple have introduced new Apple Intelligence features in iOS 18.2, iPadOS 18.2, and macOS Sequoia 15.2, including ChatGPT integration, image playground, Genmoji, and enhanced writing tools. ChatGPT allows Siri to address questions it can’t answer and supports text generation via a “compose” feature. Writing tools now let users customize edits by describing desired changes. Image Playground enables image creation in various styles, including AI remixes and animations, integrated into Messages with an API for third-party developers. Genmoji lets users create custom emojis by describing new ideas or using photos. Apple also expanded language support to localized English for regions like Canada, the U.K., and Australia. ChatGPT integration is optional, with privacy safeguards and OpenAI account login for advanced models.

Tech Trends 2025: AI's Expanding Role Unveiled

Deloitte’s Tech Trends 2025 highlights the growing integration of AI into the fabric of daily life, likening its role to foundational technologies like electricity and HTTP. AI is transforming from a standalone innovation to a critical force driving business operations, technological advancements, and societal change. The report underscores AI’s increasing role in areas like computing devices, the Internet of Things, and robotics, marking a resurgence in hardware prominence. It also focuses on AI’s ability to enhance IT workflows, including software testing and coding, while reshaping resource allocation and budgets. Furthermore, AI is simplifying user experiences while introducing greater complexity at the architectural level. The report emphasizes AI’s potential to revolutionize industries and foster future innovation.

AI Training

Master Coding Efficiency with Google’s Jules

Jules, powered by Google’s Gemini 2.0 platform, is an AI coding assistant designed to autonomously analyze and repair code. Unlike traditional tools, Jules operates as an autonomous agent within GitHub’s workflow. It reviews codebases, identifies issues, and creates detailed repair plans. Jules can implement fixes across multiple files and generate comprehensive pull requests, all while maintaining human oversight. Developers approve suggested changes, ensuring control and safety. To use Jules, integrate it into your GitHub workflow. Assign it tasks like bug fixes or code updates. Jules analyzes the code, proposes solutions, and executes changes with your approval. Developers monitor progress, review suggested pull requests, and merge updates when ready. Jules complements workflows, reducing repetitive tasks and enhancing productivity.

Heads Up 

Google unveils Trillium, its sixth-generation AI accelerator chip, promising transformative performance improvements and redefining AI development economics while advancing machine learning capabilities.

Apple partners with Broadcom to develop its first server chip for AI applications, according to reports from The Information citing insider sources.

Ori Industries becomes the first UK company to deploy Nvidia’s H200 chips, boosting capacity to meet the growing demands of AI businesses in London.

A study by Botify and DemandSphere reveals Google’s AI Overviews appear in nearly half of search results, occupying 48% of mobile screen space.

Hugging Face launches a global initiative inviting native speakers to enhance LLMs by rating educational content in their languages. No skills required, just participation!

AI Solution

Midjourney Unveils Patchwork for Storytelling Worlds

Midjourney has unveiled Patchwork, a standalone app designed for creating digital storytelling worlds. Users can build interactive, AI-generated “worlds” using an infinite canvas with tools for creating characters, places, events, and scenes. Integrated with Midjourney accounts and accessible via its Discord server, Patchwork generates visuals and text-based scraps from prompts. Users can link characters, narrate actions, and even collaborate in real-time with up to 100 contributors per world. The tool uses multiple AI models, enabling advanced storytelling and world-building, and supports downloading JSON files with generated assets. Patchwork caters to writers, game designers, and directors, with future plans for immersive 3D environments. Midjourney also announced updates, including video generation, Midjourney V7, and enhanced image personalization modes launching soon.