WednesdAI // Week 39

Innovation speeds ahead while people see warning signs.

Let’s get into it!
 
Welcome to WednesdAI – Pixel Dreams’ weekly update with top stories from the rapidly evolving world of Artificial Intelligence.

This Week’s News

Facebook Trains AI with Australian Users’ Faces and Posts

 
Facebook has admitted to scraping public posts and photos from Australian users to train its AI models. Unlike users in the European Union, Australians are not given the option to opt out. This practice includes public content from adults, though photos of children posted by adults are also collected. Meta’s global privacy director confirmed this during a Senate inquiry, prompting concerns over Australia’s outdated privacy laws. Stronger regulations, similar to the EU’s, may be necessary to protect Australian users​.
 

Discover the full information here.

Audible Cloning Audiobook Voices

 
Amazon’s Audible has launched a beta program allowing audiobook narrators to create AI-generated voice clones of themselves. Narrators record their voices, which are used to train AI models that can produce audiobooks faster and at lower cost. They maintain control over which projects use their AI voice, and can adjust elements like pronunciation and pacing to ensure quality. While narrators will earn royalties per project, there’s concern that this move could edge out human performers. The initiative is part of Amazon’s strategy to expand its audiobook catalog by tackling the backlog of non-audio titles.

Get the full scoop here.
 
 

AI Reduces Hospital Deaths

An AI system called CHARTwatch, used at St. Michael’s Hospital in Toronto, is helping reduce deaths by predicting when patients are likely to get worse. It scans tons of data—like vital signs and lab results—and alerts doctors when something looks off. Then, the human staff steps in to figure out what action to take, like moving a patient to the ICU or adjusting meds. The AI doesn’t replace doctors but acts like an extra set of (non-human) eyes, helping out when they’re overwhelmed​.

Read the full story here.
 

Suno Helps You Express Yourself in Any Style

 
Suno has launched a new feature called “Covers,” which allows users to transform songs into different styles while maintaining the original melody. Currently in beta for Pro and Premier users, the tool offers 200 free cover generations before applying a credit-based system. Suno is seeking user feedback to improve the feature, as it may have some bugs. This is aimed at giving creators more flexibility with genres and styles, opening up new creative possibilities for music production.

 
Get full details in Suno’s official release.
 
 

Videos of the Week

Exciting new releases from friends and new favorites!

The images accompanying the news items in this article were generated in Midjourney using the following prompts:

Whimsical Illustration with Playful Patterns. An enormous AI robot cheerfully scooping up clouds of tiny user photos and posts like they were candy from a jar. A bustling Australian suburb, with houses and parks below, as the robot playfully reaches down from the sky, its actions unnoticed by the residents. The AI robot is the main focus, looming large in the frame, while the clouds of user data swirl around it like colorful confetti. Bright and cheerful lighting, almost cartoon-like, with the sun shining playfully over the scene, giving the situation an ironic, light-hearted feel despite the serious subject. A couple of children laugh and play in the foreground, oblivious to the data-harvesting happening above them

Magical Realism with Bold Colors and Shapes. A narrator splitting into multiple versions of themselves, each holding a different glowing manuscript that reads itself out loud. A vibrant library with endless shelves of glowing, talking books, where each narrator clone reads in perfect synchronization with the AI versions. The central narrator is surrounded by their AI versions, each one interacting with different books as the text comes to life in the air, symbolizing how the voices bring stories into being. Bright, colorful lighting with a neon glow around the manuscripts, making the whole scene feel like a surreal explosion of sound and light. Each clone’s voice is slightly different, but they all radiate the same core energy, representing the narrator’s control over pacing and emotion in the AI-generated audiobooks

Abstract Minimalism with Dreamlike Elements. A hospital monitor that projects floating, abstract shapes representing patient health, with doctors adjusting these shapes in real-time. A minimalist hospital ICU where the AI projects holographic representations of patient health into the air, and medical staff interact with them like sculptors shaping outcomes. Doctors stand focused, moving their hands through the abstract shapes, adjusting the patterns and colors, while the AI gently shifts them in response to real-time data. Cool, soft lighting highlights the clean, sterile nature of the room, while the abstract holograms glow with warm, reactive hues. The room feels tranquil and orderly, emphasizing the harmony between AI predictions and human decision-making, with no sense of urgency or chaos

Surrealist Musical Dreamscape. A song being transformed into various visual styles, with vibrant notes morphing into different instruments and artistic elements mid-air. A fantastical floating music studio surrounded by colorful, shifting clouds where musical notes turn into abstract shapes, transforming into jazz, rock, and classical instruments as they float. In the foreground, a user interacts with an ethereal music interface, with swirling patterns of melody shifting into new genres behind them. Soft, glowing light from the instruments and notes, creating a sense of continuous flow and transformation in the misty, dreamlike atmosphere. The entire scene feels fluid and whimsical, capturing the magic of music being reinvented in real-time through Suno AI’s “Covers” tool