Welcome back to Daily AI Skills.

Here’s what we are covering today:
1. Microsoft Copilot AI for gaming
2. Google DeepMind’s Gemini Robotics Breakthrough
3. Image Generation & Editing in Gemini 2.0 Flash

+ The LLM Engineer Toolkit

Microsoft Copilot AI for Gaming

Microsoft just dropped Copilot for Gaming, an AI sidekick hitting the Xbox mobile app for Xbox Insiders in April 2025, promising to cut the fluff—less time searching, more time playing—while dishing out personalized coaching and social perks.

The Details:

• Announced on the March 13, 2025, Official Xbox Podcast, Copilot for Gaming lets you bark orders like “install Sea of Thieves” or get a recap of your last Age of Empires flop—think of it as a virtual wingman that knows your gaming habits.

• Built on capability, adaptability, and personalization, it’ll nudge you with game picks, coach you through Overwatch 2 hero choices, or ping you when friends are online—all via the free Xbox mobile app, with broader rollout planned.

• Xbox VP Fatima Kardar pitched it as a fix for gaming’s unique snag: “It’s the only entertainment where you can get stuck.” Future in-game assistance (still in concept) could offer real-time tips, but it’s optional—Microsoft swears it won’t hover like an overeager backseat gamer.

• Privacy? You’re in the driver’s seat: Opt in for tailored features or shut it off. Xbox Insiders can test it soon and shape it via feedback, starting mobile-first.

Read the Full Article by Microsoft here

Google DeepMind Unveils Gemini Robotics

Google DeepMind just unveiled Gemini Robotics, a new AI model designed to bring the power of Gemini 2.0 into the physical world by enabling robots to perform complex real-world tasks with enhanced reasoning and dexterity.

The details:

• Gemini Robotics is a vision-language-action (VLA) model built on Gemini 2.0, adding physical actions as a new output modality for direct robot control.

• A second model, Gemini Robotics-ER (Embodied Reasoning), enhances spatial understanding and real-time adaptability, allowing robots to grasp and manipulate objects more precisely.

• The models have shown a 2x–3x success rate improvement in real-world tests over previous state-of-the-art models.

• DeepMind is partnering with Apptronik to build the next generation of humanoid robots and collaborating with trusted testers like Boston Dynamics and Agile Robots.

Read the full article here

Google Adds Image Generation and Editing to Gemini 2.0 Flash

Google has introduced new experimental image-generation features for its Gemini 2.0 Flash model, allowing users to create, upload, and edit images directly within the language model — no separate image generator required.

The details:

• The 2.0-flash-exp model is now available via API and Google AI Studio, supporting both text and image outputs, with editing possible through natural conversation.

• Gemini’s multimodal reasoning enables it to maintain consistency in characters and understand real-world context throughout a dialogue.

• Users can create a story with images and refine the results by guiding the model through conversational prompts.

• Google also claims that Flash 2.0 outperforms competitors in text rendering, making it well-suited for generating ads, social posts, and other content-heavy designs.

Read the full article by Google here

The LLM Engineer Toolkit

📩Forward it to people you know who are keeping pace with the changing AI world, and stay tuned for the next edition to stay ahead of the curve!

Maybe you missed out on these:

1. Copilot vs Claude vs Gemini

2. How I Built an AI Influencer in 2 Minutes—and Landed Sponsorships

3. Microsoft Muse, Helix and Google's AI Co-Scientist

4. Bolt vs Lovable

5. Your AI Stack Just Got Superpowers: 4 New Inference Providers Live!