Sign Up | Advertise

Good morning! Today is Wednesday, April 23, 2025. We have some exciting AI news today: xAI's Grok chatbot now features real-time visual understanding, and OpenAI has expressed interest in acquiring Google's Chrome browser amid ongoing antitrust proceedings.

1. Grok Just Got Eyes: xAI’s Chatbot Can Now See Through Your Camera

Elon Musk’s xAI has supercharged its Grok chatbot with vision. The new Grok Vision feature lets iOS users point their phone cameras at real-world objects—like signs, documents, or products—and ask Grok questions about them in real time. It’s a big leap toward making AI more interactive and context-aware, joining similar features in ChatGPT and Google Gemini. Android users will have to wait, but they can now enjoy real-time voice search and multilingual audio—if they’re on the $30/month SuperGrok plan.

Read more...

2. OpenAI Wants to Buy Chrome and Turn It Into an AI-Powered Browser

In a surprising twist during Google's antitrust trial, OpenAI revealed it would jump at the chance to buy Chrome if the DOJ forces Google to sell it. With Chrome’s 4 billion users, OpenAI sees a golden opportunity to create an “AI-first” browsing experience by deeply integrating ChatGPT into every corner of the web. Backed by former Google Chrome engineers, OpenAI has already been toying with building its own browser—this could fast-track that vision and reshape how we browse, search, and interact with the internet.

Read more...

3. Oscars Say AI Is Welcome — As Long as Humans Stay at the Creative Core

The Academy Awards have officially weighed in on AI in filmmaking: it's not banned, and using it won't hurt a film’s chances at a golden statue. In a new rule update, the Oscars say what matters most is how much human creativity was involved. It’s the first time the Academy has addressed generative AI, signaling a cautious embrace of the tech as long as it doesn’t overshadow the human touch.

Read more...

4. Italy to Host Colosseum, a New Nvidia-Powered AI Supercomputer

Nvidia, Vertiv, and iGenius are teaming up to launch one of the world’s most powerful AI supercomputers — Colosseum — in southern Italy. Powered by Nvidia’s cutting-edge Grace Blackwell chips and supported by Vertiv’s modular 360AI infrastructure, the system will deliver an estimated 115 exaflops of compute power. Designed to serve highly regulated industries like finance and healthcare, Colosseum is also a symbol of Europe's push for “sovereign AI.” Thanks to Nvidia’s Omniverse, the team even built a digital twin to speed up deployment by 50%.

Read more...

5. Smarter Code, Less Waste: Researchers Use SMC to Supercharge AI Coding Tools

A team of top researchers from MIT, McGill, ETH Zurich, and more have unveiled a new technique that makes AI-generated code more accurate, efficient, and trustworthy. By adapting a method called Sequential Monte Carlo (SMC), their system guides language models to stick to programming rules and discard flawed code early in the process—saving compute power and boosting performance, especially for smaller models. In tests across Python, SQL, and even drug discovery tasks, this approach helped small models outperform larger ones, pointing to a future where AI coding tools are faster, smarter, and more reliable.

Read more...

6. Can AI Help Save Wildlife—Without Harming the Planet?

AI is rapidly transforming wildlife conservation, with researchers using drones and deep learning to track endangered species and even discover hundreds of previously unknown animals. But while tools like image recognition and acoustic monitoring speed up research, they raise big questions about environmental costs, data bias, and ethical limits. From decoding whale clicks to scanning rainforests for “extinct” plants, scientists are split: some see AI as a powerful ally in the fight against extinction, while others warn it's a double-edged sword that could do more harm than good if misused.

Read more...

Stay tuned for more updates, and have a fantastic day!

Best,
Zephyr