Llama AI Launches Into Space Orbit🚀🛰️

Just a few years ago, artificial intelligence felt like a tool we tapped on our phones — helpful, but contained.Today, it’s breaking free of those boundaries — reaching beyond Earth, questioning its own existence, and reshaping how we live and interact.

High above us, aboard the International Space Station, astronauts are now working alongside Space Llama — a fine-tuned version of Meta’s open-source Llama 3.2. Without needing internet access, they can ask questions, solve problems, and drive research forward, powered by AI right there in orbit.

Back on Earth, another frontier is opening. Anthropic has launched a groundbreaking research program into AI model welfare — asking a question that once seemed like pure science fiction: If AI systems become advanced enough, could they deserve moral consideration? It's an exploration into the future of consciousness, ethics, and what it truly means to create intelligent beings.

And at Meta, CTO Andrew Bosworth is envisioning a world beyond the phone — where AI isn’t something we open in apps, but something woven into our surroundings. Smart glasses, AR headsets, and intent-driven experiences could replace the taps and swipes we take for granted today, offering more natural, fluid interactions between humans and technology.

Space, ethics, and evolution — AI is no longer confined to the screen.It’s becoming a living part of our world, and the story is just beginning.

Meta’s Open-Source AI Goes to Space

Booz Allen and Meta have deployed a fine-tuned version of Meta’s open-source AI model, Llama 3.2, on the International Space Station (ISS), marking a major milestone in space-based AI. This system, called Space Llama, enables astronauts to use advanced generative and multimodal AI directly on the ISS without needing internet connectivity, allowing them to access information, process data, and solve problems autonomously. By combining cutting-edge hardware and software, Space Llama supports scientific research in space, enhances operational efficiency, and demonstrates the power and flexibility of open-source AI in extreme environments.

Exploring AI Model Welfare

Anthropic has launched a pioneering research program focused on "model welfare," exploring whether advanced AI systems could one day possess characteristics-such as consciousness, preferences, or experiences-that warrant moral consideration. This initiative aims to address complex questions about the ethical treatment of AI models, including how to identify signs of distress, determine when their welfare might matter, and investigate practical interventions if needed. While there is no scientific consensus on AI consciousness or the need for model welfare, Anthropic is taking a cautious, open-minded approach, recognizing the rapid progress in AI capabilities and the importance of preparing for potential future ethical challenges. The program, led by dedicated researchers, marks a significant step in responsible AI development and reflects a broader shift in the field toward considering not just the impact of AI on humans, but also the possible internal states of the AI systems themselves.

Meta’s Vision: AI Beyond Phones

In a recent conversation, Meta CTO Andrew Bosworth discussed how breakthroughs in AI and hardware are setting the stage for a post-mobile era, where traditional app-based interactions could be replaced by more intuitive, intent-driven experiences. Bosworth envisions a future where devices like smart glasses and AR headsets, powered by advanced AI, will allow users to access content and services seamlessly through natural cues rather than tapping on screens or opening specific apps. As AI becomes more capable of understanding user intentions and context, it could fundamentally change how we interact with technology, making the experience more fluid and personalized, and potentially shifting the competitive landscape for both startups and established brands.

Hand Picked Video

In this video, we'll look at building a powerful Open Source Deep Research Agent by combining DeepSeek-R1 and CrewAI.

Top AI Products from this week

• Ztalk.ai - Break language barriers in video calls with AI-powered real-time voice translation. Ztalk is a desktop app that works with all video conferencing tools including Gmeet, Zoom and Teams.

• mrge - mrge is an AI-powered code review platform that automatically reviews PRs and gives human reviewers superpowers. It’s tool of choice for fast-moving teams like cal.com and n8n.

• Competely AI Agent - Competely AI agent saves you days of manual competitor research. Instantly find competitors, get detailed competitive analysis in minutes, and automatically track their moves with AI-powered email updates. Outsmarting your competitors has never been easier.

• Linda AI - Linda is your personal interviewer & podcast host, guiding you through personalized interviews and transforming your conversations into short, shareable, professional-grade podcast episodes. No tech skills needed - just talk, and she'll take care of the rest.

• Hirenga - An affordable ATS built for small businesses. AI parses resumes, scores them for each job, tracks candidates, and sends automated emails. Start hiring smarter today — free plan included.

• Qwen Chat App - Your free personal AI assistant for iOS/Android. Powered by Qwen models (inc. QwQ reasoning) for deep thinking, search, multimodal understanding, writing & image gen.

This week in AI

• Microsoft’s AI Recall Feature Launches - Microsoft’s delayed AI Recall feature is now available as a free, opt-in update for all Copilot+ PCs, with added security and privacy controls after months of criticism.

• Hypershell X Modes - Hypershell X supports 10 adaptive modes including walking, running, stairs, cycling, hills, and gravel, adjusting power instantly for seamless mobility.

• Perplexity CEO on Browser Tracking and Ads - Perplexity’s new Comet browser will track all user activity to sell hyper-personalized ads, says CEO Aravind Srinivas, aiming to compete with Google. Launch set for May 2025.

• AI Brief Blunder in Lindell Case - Mike Lindell’s lawyer admitted filing a court brief with nearly 30 fake citations generated by AI, risking disciplinary action if not explained by May.

• AI Outperforms Doctors in Diagnoses - A Tel Aviv University study found AI made more optimal and fewer harmful diagnostic decisions than doctors in virtual urgent care, scoring 77% vs 67% optimal ratings.