A recent study found that an AI chatbot-powered software company could develop software in under seven minutes for less than $1 in costs, on average. ILYA SEDYKH/Getty.
Category: robotics/AI – Page 872
System combines light and electrons to unlock faster, greener computing
Computing is at an inflection point. Moore’s Law, which predicts that the number of transistors on an electronic chip will double about every two years, is slowing down due to the physical limits of fitting more transistors on affordable microchips. Increases in computer power are slowing down as the demand grows for high-performance computers that can support increasingly complex artificial intelligence models.
This inconvenience has led engineers to explore new methods for expanding the computational capabilities of their machines, but a solution remains unclear.
Photonic computing is one potential remedy for the growing computational demands of machine-learning models. Instead of using transistors and wires, these systems utilize photons (microscopic light particles) to perform computation operations in the analog domain.
AI training models will be 1,000x larger in three years
If you thought ChatGPT was impressive, you ain’t seen nothing yet…
DeepMind co-founder Mustafa Suleyman predicts ongoing, exponential progress in LLMs and other generative AI. But the emergence of such powerful technology raises huge ethical and safety concerns.
DeepMind co-founder Mustafa Suleyman predicts that AI will continue its exponential progress, with orders-of-magnitude growth in model training sizes over the next few years.
A.I. Sampling and how the Music Industry will change forever
If people remember how sampling changed music, watch what this guys does to make AI music. A long time ago when people said AI will replace musicians, I replied AI is just a sampler. If people use a Tupac voice on a song like this guy did, they just pay royalties. Then with samplers arists made sample disks royalty free. They make money when you buy the sample disk. The same with AI, you just upload your sample disk into your AI, whether the music AI is from Meta or Google. Yeah Meta has music AI, you can see it used here.
Welcome to a showcase of sounds sampled through the power of artificial intelligence. Gone are the days of vinyl digging; now, we embrace prompt digging…
Jump on the hype train for this channel, and help me crank out even more wicked videos like this one:
https://www.patreon.com/NobodyandTheComputer.
Contact me:
[email protected].
Colab Notebook META AudioGen & MusicGEN:
These personal robots can explain emotions in many languages
Jibo, a robot engineered by MIT Professor and Dean for Digital Learning Cynthia Breazeal, who is also the founder of the institution’s Media Lab Personal Robots Group (PRG), is a personal robot that can actually do what even few humans can: explain emotions. It achieves this in several languages, including English and Arabic.
Move over AI, quantum computing will be the most powerful and worrying technology
Head over to our on-demand library to view sessions from VB Transform 2023. Register Here
In 2022, leaders in the U.S. military technology and cybersecurity community said that they considered 2023 to be the “reset year” for quantum computing. They estimated the time it will take to make systems quantum-safe will match the time that the first quantum computers that threaten their security will become available: both around four to six years. It is vital that industry leaders quickly start to understand the security issues around quantum computing and take action to resolve the issues that will arise when this powerful technology surfaces.
Quantum computing is a cutting-edge technology that presents a unique set of challenges and promises unprecedented computational power. Unlike traditional computing, which operates using binary logic (0s and 1s) and sequential calculations, quantum computing works with quantum bits, or qubits, that can represent an infinite number of possible outcomes. This allows quantum computers to perform an enormous number of calculations simultaneously, exploiting the probabilistic nature of quantum mechanics.
Apptronik Unveils New Humanoid Robot, Apollo
Apptronik, an Austin-based robotics start-up, has revealed its latest humanoid robot, Apollo. Standing at 5 feet 8 inches tall and weighing 160 pounds, Apollo is designed for mass production and safe human-robot collaboration. Unlike traditional robots, Apollo uses electricity instead of hydraulics, making it both safer and more efficient.
Apollo is equipped with a four-hour battery life that can be easily exchanged for continuous use up to 22 hours, allowing it to perform physically demanding and dangerous tasks, improving supply chains and reducing human risk.
To ensure that Apollo is accessible and friendly, Austin-based company Argodesign has equipped the robot with features such as digital panels on its chest for clear communication, intentional movements like head rotation, and a friendly face.
Tesla (TSLA) stock surges from optimistic look at Dojo supercomputer
Tesla’s (TSLA) stock is rising in pre-market trading on an optimistic new report about the automaker’s Dojo supercomputer coming from Morgan Stanley.
The firm massively increased its price target on Tesla’s stock because of it.
Dojo is Tesla’s own custom supercomputer platform built from the ground up for AI machine learning and, more specifically, for video training using the video data coming from its fleet of vehicles.
Meta’s AI Agents Learn Via Toddler-Like “Motor Babbling”
Similarly, allowing the MyoLegs to flail around for a while in a seemingly aimless fashion gave them better performance with locomotion tasks, as the researchers described in another paper presented at the recent Robotics Science and Systems meeting. Vittorio Caggiano, a Meta researcher on the project who has a background in both AI and neuroscience, says that scientists in the fields of neuroscience and biomechanics are learning from the MyoSuite work. “This fundamental knowledge [of how motor control works] is very generalizable to other systems,” he says. “Once they understand the fundamental mechanics, then they can apply those principles to other areas.”
This year, MyoChallenge 2023 (which will also culminate at the NeurIPS meeting in December) requires teams to use the MyoArm to pick up, manipulate, and accurately place common household objects and to use the MyoLegs to either pursue or evade an opponent in a game of tag.
Emo Todorov, an associate professor of computer science and engineering at the University of Washington, has worked on similar biomechanical models as part of the popular Mujoco physics simulator. (Todorov was not involved with the current Meta research but did oversee Kumar’s doctoral work some years back.) He says that MyoSuite’s focus on learning general representations means that control strategies can be useful for “a whole family of tasks.” He notes that their generalized control strategies are analogous to the neuroscience principle of muscle synergies, in which the nervous system activates groups of muscles at once to build up to larger gestures, thus reducing the computational burden of movement. “MyoSuite is able to construct such representations from first principles,” Todorov says.