Wingin' it

Good morning Staker! Here’s what’s cookin’ today: Consumer sentiment hits another historic low; Scientists share one simple hack for managing weight loss; AI has cloned all of Wikipedia; and The Rolling Stones turn back the clock in the video for their new release.

But first, our weekly poll!

A study conducted by researchers with the American Psychological Association suggests adhering to a diet that includes the same foods and same caloric intake every day can lead to increased weight loss.

Published in Health Psychology, the study tracked participants for a period of 12 weeks, with one group eating mostly the same meals each day and consuming the same amount of calories, and the other group frequently mixing up their foods and calorie count.

The researchers found the first group achieved more weight loss than the second group, despite similar exercise and sleep patterns.

"Maintaining a healthy diet in today's food environment requires constant effort and self-control," said lead author Charlotte Hagerman, PhD, of the Oregon Research Institute. "Creating routines around eating may reduce that burden and make healthy choices feel more automatic."

Specifically, the researchers found the consistent meal group achieved 5.9% weight loss compared to 4.3% for the other group. Each additional 100 calories consumed also resulted in 0.6% less weight lost.

"If we lived in a healthier food environment, we might encourage people to have as much variety in their diet as possible," Hagerman said. "However, our modern food environment is too problematic. Instead, people may do best with a more repetitive diet that helps them consistently make healthier choices, even if they might sacrifice some nutritional variety."

There’s a whole lot going on physiologically when we speak, even just when it comes to the coordination of breathing and speaking at the same time.

This particular example of multitasking is taken for granted most of the time, but that coordination becomes a lot more apparent when we’re walking quickly or running, and a handful of vocal characteristics are impacted.

"Physical exertion directly alters respiration and phonation, and because speech shares the same respiratory system, these changes propagate into pitch, timing, and voice quality," said Zahra Omidi of the University of Texas at Dallas, who studies this exact relationship.

Omidi is working on mechanizing the characteristics of a tired voice to improve various types of speech pathology technology that’s used for helping people recover from injuries that impact speech and when humans work in conditions under duress.

"Examples include emergency response and military operations, and wherever people are speaking while physically active," Omidi said.

"In all these cases, speech deviates from neutral conditions due to respiratory and vocal effort constraints, leading to reduced intelligibility and system performance."

Remember the other day when Stake discussed the concept of entropy, and how it can be used to describe the aging process, but is generally a physics term that refers to the descent of all things towards disorder?

We just came across another perfect example of entropy, and it comes in the form of Halupedia.

It’s a website that somewhat resembles Wikipedia, but its entire contents are made up of AI-generated hallucinations. It’s an encyclopedia devoted solely to cataloguing a level of nonsense scholars and computer programmers alike wouldn’t have been able to fathom even 20 years ago.

Basically, every article on every topic that anyone has ever prompted the Halupedia LLM to compose is available to read. Take, for instance, this entry on the Great Leaf Dispersal of 1732. This entire article is complete and utter nonsense.

While entertaining, sites like this actually paint a good picture of where many believe the internet is headed, thanks to the ongoing and endless churning out of content by LLMs.

When AI models were first built, they were trained on all the human-generated information that the internet had to offer. Every time an LLM generates content, that content also becomes part of the pool of information AI models are trained on moving forward.

In other words, the quality of content has been eroding since the minute LLMs launched, and as their training materials descend further into nonsense and they pump out more content than any human can keep up with, some believe the internet is destined to become a graveyard of unintelligible AI-generated slop. 🤯

The Rolling Stones released the music video for In The Stars, one of two singles off the band’s upcoming album Foreign Tongues, and in the video, they don’t look a day over 30.

That’s not an exaggeration or a commentary on how incredible it is that this band continues to ooze rock ‘n roll 62 years into their career. They don’t look over 30 because the video utilizes what director Francois Rousselet and creator Deep Voodoo describe as "groundbreaking deepfake technology.”

Does anyone really want AI anywhere near music? Not really. Is it cool to see it used to turn back time in a portrayal of the Rolling Stones in their prime? Sure.

The Stones themselves have weighed in previously with their thoughts on the rising tide of AI in music.

"If you really want to listen to a record properly...I mean digital is toy town," Keith Richards said in 2023 on an episode of Sidetracked with Annie [Mac] and Nick [Grimshaw] (via NME).

"Synthesizers, now you have AI, which is even more superficial...But AI is like anything else. It can either be a tool, or it can be a toy. And most times, all of these things become toys. But it's how you use it."

Quizmaster’s choice

It’s another Friday, and that means it’s time for another game of à la carte for today’s trivia! 😏