Apple is Building a Pocket Sized AI Health Coach

Midjourney just dropped its V7 AI image model, and it’s packed with amazing upgrades. One of its major highlight is the Draft Mode, which lets you generate images 10 times faster at half the cost. Perfect for brainstorming, it gives out rough drafts you can later polish. But the real game-changer is voice prompting, just talk to the AI, and it turns your rambles into art. Imagine saying, “Make this darker with more texture,” and watching the image transform in real-time. Hands and textures, AI’s longtime weaknesses, now look shockingly real, though some users say it’s more of a “V6.2” than a full leap. To use V7, you will need to rate 200 images to build a personal style profile, which the AI uses to tailor creations to your taste. Missing features like upscaling are coming soon, but for now, Turbo Mode (fast but pricey) and Relax Mode (slow but detailed) are your options. Mixed reviews aside, it is a fun step toward seamless human-AI teamwork.

Apple is working on an AI health coach codenamed Project Mulberry, set to launch in 2026. It will live inside a revamped Health app (possibly called Health+) and act like a personalized trainer. Using data from your Apple Watch and iPhone, it will nudge you to sleep better, eat healthier, and even fix your workout form using the phone’s camera. The AI analyzes heart rate, activity, and meal logs to give tailored advice—think “Eat more protein” or “Try a 10-minute mindfulness session.” Apple is collaborating with doctors to ensure accuracy, and users get access to expert videos on wellness topics. Rumor has it some features might require a subscription, aligning with Apple’s push into services. While it is not a replacement for doctors, it’s a bold step toward proactive healthcare and another sign Tim Cook is serious about making Apple a health tech titan.

Amazon just unveiled Nova Act, an AI agent that can browse the web and complete tasks like shopping, booking appointments, and filling forms just like a human would, except it works autonomously without needing constant supervision, using "atomic commands" to break complex tasks into smaller, more reliable steps while securely handling sensitive data through integration with Playwright, and it's already being tested in features like "Buy for Me" where it can purchase items from third-party sites directly through the Amazon app, with the company claiming it outperforms competitors like OpenAI's Operator and Anthropic's Claude in following on-screen instructions accurately, though some critics remain skeptical since Amazon hasn't used standard benchmarks, while developers can already experiment with the free Nova Act SDK to build custom automation workflows, marking Amazon's big bet that practical AI agents that do things rather than just chat will be the next computing revolution.

Meta recently introduced MoCha, an AI system that transforms simple text scripts into fully animated characters with natural body movements, facial expressions, and perfectly synced lip movements, going beyond previous AI animation tools by generating full-body performances without motion capture, using a novel "speech-video window attention" technique to maintain precise lip-sync and allowing creators to define and reuse characters across multiple scenes, producing high-quality 5-second video clips at 24fps that testers rated as remarkably lifelike, positioning it as a potential game-changer for content creation, education, and virtual influencers, though Meta hasn't revealed when or how the public might access this technology as it joins an increasingly competitive AI video generation space alongside tools from ByteDance and Runway, raising questions about how such technology might reshape entertainment and digital communication in the future.

Podcastle is designed to let long-form content creators do what they do best, from stunning videos to compelling podcasts to lifelike AI audio.

Stay Tuned! We will update you as soon as a recent opportunity is available