Warp-Speed Wednesdays
Your Must-Read Weekly Tech Updates: Artemis V Moon Mission: Back to the Moon to Stay, AI reads your mind, AI Regulatory Talks, and more Medical AI Breakthroughs
Click that weblink here to get the full galactic experience.
Space
We are going to the Moon… to STAY!!!!
NASA has officially teamed up with Blue Origin, and they're about to take our lunar explorations to heights that would make even Neil Armstrong's jaw drop.
It's not just any mission, either. We're talking about Artemis V — an expedition that will see more of the moon than ever before, paving the way for future Mars voyages and filling our heads with more sci-fi dreams than you can shake a lunar rock at.
Blue Origin is gearing up to design, develop, test, and verify its Blue Moon lander, a piece of tech so advanced it's like something out of a science fiction novel. And what's the price tag on this cosmic adventure? A cool $3.4 billion!
What makes this particularly thrilling is the uncrewed demonstration mission, which will be followed by a manned mission on Artemis V in 2029. Yes, you heard that right — we're going to be setting boots on the moon again in just a few short years!
But wait, there's more! Artemis V is set to make a weeklong trip to the Moon's South Pole region for science and exploration activities. It's the epitome of lunar exploration and the first step towards recurring complex missions as part of NASA’s Moon to Mars exploration approach.
And what's the cherry on top of this celestial sundae? The addition of another human landing system partner in NASA's Artemis program. This not only increases competition and reduces costs to taxpayers but also fuels the lunar economy and sets the stage for future astronaut missions to Mars.
We're in the golden age of human spaceflight, and it's about to get a whole lot more exciting!
Also, here is NASA’s blueprint to go from the Earth to the moon to Mars.
“We choose to go to the moon, in this decade, and do the other things. Not because they are easy, but because they are hard. Because that challenge is one that we are willing to accept. One that we are unwilling to postpone.”
President John F. Kennedy
Artificial Intelligence
AI Regulation Talk Show and the Unspoken Power of Palantir’s Military AI.
While there's been a buzz around Sam Altman and other AI mavens conferring with Congress about regulation, I must express my reservations about the whole affair. At first glance, it feels like a performance, orchestrated more for optics than actual action, especially when Palantir's Alex Karp and Peter Theil are not also at this congressional meeting. For you who aren’t aware, Palantir is presently perfecting a military AI for the US government. Yes, this technology holds significant potential to protect American lives, an undeniably noble endeavor. Yet, we must pause to consider the larger picture — the cost to human lives outside American borders. We need to ask ourselves these hard questions as we navigate the complex terrain of AI deployment.
Alex Karp, CEO of Palantir, says this about his company’s technology:
"Which of our special forces units are closest to enemy tank positions and have sufficient supplies of Javelin missiles to mount an offensive? And which specific tanks on the battlefield are most vulnerable to attack?"
"The applications of these newest forms of artificial intelligence have been and will continue to be determinative on the battlefield.”
"Others can debate the merits of proceeding with the development of these technologies. But we will not stand still while our adversaries move ahead."
Palantir is one of the few companies, if not the only one, that truly sits at the intersection between public, private, and government data. The power they wield is truly immense!
Karp reminds us of their power by stating,
"The union of public datasets, such as the corpus of text and information available on the internet, with privately held repositories of information maintained by government agencies and companies, will transform the latest large language models into something more than objects of popular fascination," Karp writes. "This emerging group of technologies will ultimately allow not only thousands but hundreds of thousands and even millions of users to interact with and manipulate datasets that until now have been functionally invisible to most people."
Unleashing the Power of 'Tree of Thoughts' in Language Models.
Artificial language agents just had a major upgrade with the advent of Tree of Thoughts (ToT), a breakthrough framework that promises to redefine how we view problem-solving in the world of AI.
Picture this — language models stepping out of their token-level, left-to-right decision-making boundaries, and into a space where they can look ahead, evaluate multiple paths, and make globally considered decisions. Sounds like sci-fi, right? Not anymore!
ToT is here to make it a reality by introducing a more deliberate decision-making process. Attention all code whisperers and robo-geeks! It's time to dust off those algorithms textbooks. The authors used breath-first search and depth-first search as their search architectures. These algorithms are fundamental, and we should see more advanced search algorithms used soon, like Monte Carlo and others, based on these amazing results.
These search algorithms allow language models to backtrack, evaluate different reasoning paths, and decide the best next step — very much like a human would in problem-solving situations!
To say the results have been impressive would be an understatement. The ToT approach blew us away during testing on tasks requiring advanced planning or search capabilities. On the Game of 24, for instance, while the previous method solved a mere 4% of tasks, ToT stole the show with a whopping 74% success rate!
Hold onto your neural networks, folks, because science just took a giant leap forward! We're talking about Mind-Video, a ground-breaking project that reconstructs videos directly from human brain activity. Yes, you read that right — it's like literally watching your thoughts on a screen! Now, while the tech isn't flawless, this is an epic advancement from just last year when creating consistent live images seemed like a far-off dream.
The magic lies in Mind-Video's ability to learn from continuous fMRI data, bringing your cerebral cortex to the big screen. And the results are impressive — they are seeing an 85% accuracy in semantic classification tasks and a 0.19 structural similarity index (SSIM), which is a massive 45% leap beyond previous methods.
What makes it even more exciting is its biological plausibility and interpretability — it mirrors established physiological processes. It's not just science; it's practically a sci-fi thriller unfolding in the lab!
Medicine
The world of medical AI is witnessing an exciting breakthrough! Med-PaLM 2, a new version of the AI system for answering medical questions, is making waves. This improved version scores a whopping 86.5% on the MedQA dataset, marking a 19% improvement over its predecessor. It's even managing to win over physicians! In comparative rankings, doctors preferred Med-PaLM 2's responses over their own on eight out of nine relevant clinical axes. While further real-world testing is needed, these findings signal a promising leap towards AI matching physician-level expertise in medical problem-solving. How exciting is it to imagine the future of medical AI?!
This new AI tool predicts Parkinson’s disease with 96% accuracy up to 15 years in advance. Presenting an innovative neural network (NN) framework that makes sense of entire metabolomics datasets without pre-selected features, this approach accurately predicts diseases like Parkinson's with a success rate that surpasses existing ML methods. The framework not only successfully identifies Parkinson's disease with a mean accuracy of over 99.5%, but also pinpoints specific markers, including an exogenous poly-fluoroalkyl substance, that show up even before clinical diagnosis. This breakthrough in 'omics' methods paves the way for remarkable advances in disease diagnostics.
Incoming Galaxy!!!!
Coming this Friday, May 26th. Introducing a new incoming Tech Galaxy “Don't Panic!: Everything Guides”. Keep your eyes peeled for the first planet “From AI Padawan to Jedi Master: The Mostly Harmless AI Cheat Sheet”. This cheat sheet will have everything you need to know about AI — from Novice to AI Researcher — and will be continuously updated. Your ever-expanding Universe awaits!
You will see a tab on the Home page when the first guide is launched.
Don’t worry!.. You will get notified when the guides are released… if you’ve subscribed. It’s free!