This week was relentless. Just when I thought the AI design scene couldn't move faster - it did. In fact, it sprinted ahead, full-force, and crashed through my office wall like it was the Koolaid Man.

First, Higgsfield drops cinematic AI that turns a single image into a Hollywood-level video (seriously, Spielberg, watch your back).

Then ElevenLabs rolls out voice cloning from just a text prompt, so now even Melania Trump is using it to narrate her memoir (you literally cannot make this stuff up).

And just as my brain was about to short-circuit, FLUX.1 Kontext appeared with hyper-realistic context-aware image editing.

Honestly, I'm struggling to keep up. But here's the good news: designers, we've never had tools this powerful, this smart, and this accessible. Our creative process just got a major upgrade.

Now, if only we had a design tool making advancements at the same pace as these AI breakthroughs. Enter Magic Path...

From design to live in seconds.

Focus on what it looks like live, we’ll do the rest. Publish your design as a live site in seconds. Hit publish, and your site is instantly live worldwide on state-of-the-art hosting. Need to upgrade something? Just as fast.

Get 25% off for 3 months with code TODAYINDESIGN

Watch the interview here 👇

(P.S. Our thumbnail experiments are hilarious)

Magic Path wants to create a new design workflow

When Pietro first showed me Magic Path, I immediately thought, "Oh man, this is exactly what my brain wishes Figma could do." The way it effortlessly spins up fully interactive designs from simple prompts is the kind of designer superpower I've been waiting for. It’s less about being an "AI design tool" and more about finally having that mythical intern who reads your mind (without the awkward coffee runs).

Watching Pietro work was like catching a glimpse of what the next few years of design might actually look like - minus the terrifying robo-dystopia. He’s not just chasing AI hype; he's genuinely focused on crafting a workflow where you can quickly explore crazy ideas, refine them, and ship them without ever getting bogged down in backend headaches. Honestly, it's refreshing to see someone who still believes designers should actually have fun at work.

Also - and I say this with deep envy - Magic Path is clearly built by someone who’s spent way too many nights obsessing over pixel-perfect details. The subtle animations, thoughtful interactions, and frankly unfair level of polish made me want to simultaneously applaud and ask Pietro if he ever sleeps. It's like discovering your friend secretly owns a Ferrari after you've spent years carpooling in a Prius.

👉 A quick favor

1. Watch: This Tool Changes Design Forever, Pietro Schirano, Magic Path

2. Subscribe: New designer-turned-founder chats drop 4x a month—pure signal.

HIGGSFIELD

Higgsfield—Stable Diffusion’s New Control Room

Stable Diffusion’s notoriously wild creativity finally has a grown-up in the room: Higgsfield. Think of it as mission control for your diffusion models, letting you easily manage embeddings, LoRAs, and checkpoints without descending into command-line chaos.

Early users swear by its tidy UI and intuitive versioning, calling it “the Figma of image model management.” If you’ve ever accidentally nuked your perfect anime character with a rogue setting, this is your new favorite safety net.

FLUX

Flux.1 Kontext just replaced Photoshop 🤯

If you've ever wrestled with a generative AI that confidently changes everything except the one tiny detail you asked for - meet your new best friend, Kontext from Black Forest Labs. Instead of playing prompt roulette and praying to the masking gods, you just tell Kontext plainly what you want. Change the car to blue, swap out some awkward text -bam, done. Everything else stays exactly where you left it. Sorcery.

But my favorite part? Kontext isn't just obedient - it's consistent. You can keep tweaking an image repeatedly, and it'll remember exactly what you meant by "make the puppy cuter," rather than inventing some terrifying creature every iteration. If you care about the tiny details (like me, obsessively), it actually respects your creative vision without drifting off into the AI abyss.

APPLE DESIGN AWARDS

Play Wins Apple Design Award for Innovation - And We’re Not Even a Little Surprised

Apple just handed out its 2025 Design Awards, and in a move that shocked absolutely no one in our Slack, Play took home the Innovation trophy. If you’ve ever used Play, you know why: it’s the rare design tool that doesn’t just talk about bridging design and development - it actually does it. You design with real SwiftUI components, prototype directly on your iPhone, and then ship straight to Xcode. It’s like someone finally decided to build a design tool that respects both your time and your taste.

What makes this win even sweeter is that Play isn’t some faceless mega-corp - it’s a small, scrappy team that’s been laser-focused on making mobile design feel native, fluid, and dare I say… fun? They’ve been one of our favorite sponsors for a reason: they get it.

ELEVENLABS

Designing voice experiences just got a major unlock

ElevenLabs just dropped their Multilingual v3 model, and suddenly the AI voice world feels less robotic and way more cosmopolitan. With native-sounding fluency in over 30 languages, it’s like ElevenLabs sent their models on an extended backpacking trip across Europe—and they came back cultured and annoyingly well-spoken. It’s now disturbingly easy to make your AI say “bonjour,” “hallo,” or “ciao” without sounding like a tourist fumbling through a phrasebook.