This issue is brought to you by tastytrade a platform designed for people who take trading seriously.

No hype. No shortcuts. No watered-down experiences.

Just the tools that matter.

Trade stocks, options, futures, and more, all in one place, with zero commission on stocks.

Use AI-powered search to surface opportunities that actually align with your strategy.

Dive into advanced charting, backtesting, and risk analysis built for real decision-making.

When speed matters, there’s Active Trader Mode and one-click order entry.

When precision matters, there’s smart order tracking, real-time data, and a fully customizable workspace.

And when you want to get better not just busier you’ll find expert-led courses and live Trade Desk support throughout the trading day.

Award-winning for a reason.

If you’re ready for a platform that matches your ambition:

tastytrade.com/Unleashed

"Quick question: Who are you? 🤔

I want to get a better pulse on the roles, industries, and interests that make up our community. If you have 60 seconds to spare, filling out this quick survey will directly shape what we do next!"

If you wear glasses or contacts like I do, you've probably felt a bit left out of the smart glasses trend. They always seemed built for people with perfect vision who just wanted cool sunglasses. Well, Meta just dropped their first prescription-optimized AI glasses—the Ray-Ban Meta Blayzer and Scriber Optics. They're designed from the ground up for people who need to wear their glasses all day, which means actual, breathable comfort.

They're rolling out hands-free nutrition tracking. You literally just look at your food, ask Meta AI to log it, and it pulls the nutrition details for you. Plus, they're introducing "Neural Handwriting," which lets you write in the air with your finger to reply to messages silently on WhatsApp, iMessage, and Instagram. We are officially in sci-fi territory here. They start at $499 and hit stores in mid-April.

Google just launched Veo 3.1 Lite. If you've been wanting to build apps with AI video generation but found the API costs way too steep, this is huge news. It’s less than half the price of their Veo 3.1 Fast model, but it runs at the exact same speed.

It supports both text-to-video and image-to-video in 1080p and 720p, and you can seamlessly switch between landscape (16:9) and portrait (9:16) framing. Basically, it’s Google’s way of democratizing AI video so developers can start baking it into their apps without going bankrupt on compute costs.

Inception Labs just released Mercury Edit 2. This is a diffusion-based LLM built specifically for "next-edit prediction" in your code editor.

Because it uses a diffusion model to generate code tokens in parallel rather than one by one, it's insanely fast. It analyzes your recent edits and codebase context to predict exactly what you're about to change next. The latency is so low it practically feels like it's a part of your own thinking process. You just hit Tab to accept the edit. If you use the Zed editor, they even dropped a promo code in their announcement for a free month to test it out.

FIPO: Eliciting Deep Reasoning with Future-KL Influenced Policy Optimization

AI models often struggle with complex, multi-step logic problems. FIPO fixes this by changing how the AI is rewarded during training. Instead of treating every generated word equally, it gives more value to crucial words that shape the final correct answer, helping the model think much longer and smarter.

Unify-Agent: A Unified Multimodal Agent for World-Grounded Image Synthesis

Unify-Agent makes generating realistic images smarter by treating the AI like an active researcher. Instead of simply drawing from a basic prompt, the AI searches for factual evidence, rewrites its instructions based on reality, and then creates the image. This step-by-step process produces highly accurate, real-world images. 

VGGRPO: Towards World-Consistent Video Generation with 4D Latent Reward

AI-generated videos often suffer from warped shapes or jittery cameras. VGGRPO solves this by directly teaching the AI the rules of 3D space. By rewarding smooth camera movements and consistent physical structures without relying on slow, heavy calculations, it creates stable, high-quality videos that look physically correct.

Z.ai Unveils GLM-5V-Turbo Vision Coder Z.ai has launched GLM-5V-Turbo, a smart computer program that can look at visual designs—like website layouts or app screenshots—and instantly write the code to build them. The system operates like a helpful digital assistant that can "see" your screen and complete complex tasks, all without losing its core text-based programming skills. It achieves this advanced performance by being trained from the ground up to seamlessly blend visual understanding with logical problem-solving.

Apple CarPlay ChatGPT Update OpenAI has updated its popular chatbot app to work directly on car dashboard screens through Apple's CarPlay system. Drivers can now talk to the smart computer program completely hands-free while driving, keeping their eyes safely on the road. This makes it the first major voice assistant of its kind to officially work with Apple's built-in car display system.

OpenAI's Next Phase OpenAI recently raised $122 billion to build the massive computer systems needed to power the future of artificial intelligence, which refers to computer programs designed to think and learn like humans. The company plans to combine its various tools into a single, simple application that can understand what users want and automatically complete tasks across different programs. This huge amount of money will help them handle the growing global demand for these smart computer helpers in both daily life and business.

Oracle Layoffs Oracle is firing about 18% of its workers, affecting up to 30,000 people, to save money for massive investments in advanced computer technology. Instead of using smart computer programs to do the jobs of the fired employees, the company is simply eliminating roles to pay for expensive new projects and giant rooms full of computer servers. This harsh move shows how much pressure large technology companies feel to spend heavily to keep up with new trends.