Google just launched Veo 2, their next-level AI video model, for users in the MENA region—and yes, we jumped on it the moment we got early access.

Naturally, our first experiment? A classic Egyptian man drinking his tea, because we are El-Shai, after all. If we’re testing Google’s latest AI video tool, it only makes sense we start with something that literally has our name written all over it.

So, What’s Veo 2?

It’s Google’s most advanced video AI to date, now integrated into their Gemini platform. Think of it as ChatGPT’s video-savvy cousin. Veo 2 turns text prompts into short, high-quality 8-second videos.

The results? Surprisingly realistic—cinematic motion, natural lighting, and details that actually reflect how people move and how the real world works. It even gets human gestures right, which is more than we can say about a lot of other tools we’ve seen.

Is it perfect? No AI video tool is 100% there yet—and we’re not pretending this one is. But is it the closest thing to reality we’ve tried so far? Absolutely.

Here’s what makes it interesting:

• It generates 720p MP4 videos in landscape format.
• Each frame includes a SynthID watermark, so there’s full transparency that it’s AI-made.
• It works in both Arabic and English (finally, some proper regional language support!).
• And yes, you can access it through Gemini on both web and mobile—just make sure you’re a Gemini Advanced subscriber.

Veo 2 is now available globally, and Google is rolling it out across platforms in the coming weeks. We’re already thinking about what to try next—Alexandria sunsets? Cairo traffic chaos in 8 seconds?

If you’re using Gemini Advanced, head to gemini.google.com and give it a try. Just maybe don’t start with anything too weird. Or do. We’re not here to judge.