Toggle light / dark theme

Google DeepMind Unveils Veo 2, A New AI Video Model To Rival OpenAI’s Sora

The notion of entropy grew out of an attempt at perfecting machinery during the industrial revolution. A 28-year-old French military engineer named Sadi Carnot set out to calculate the ultimate efficiency of the steam-powered engine. In 1824, he published a 118-page book(opens a new tab) titled Reflections on the Motive Power of Fire, which he sold on the banks of the Seine for 3 francs. Carnot’s book was largely disregarded by the scientific community, and he died several years later of cholera. His body was burned, as were many of his papers. But some copies of his book survived, and in them lay the embers of a new science of thermodynamics — the motive power of fire.

Carnot realized that the steam engine is, at its core, a machine that exploits the tendency for heat to flow from hot objects to cold ones. He drew up the most efficient engine conceivable, instituting a bound on the fraction of heat that can be converted to work, a result now known as Carnot’s theorem. His most consequential statement comes as a caveat on the last page of the book: “We should not expect ever to utilize in practice all the motive power of combustibles.” Some energy will always be dissipated through friction, vibration, or another unwanted form of motion. Perfection is unattainable.

Reading through Carnot’s book a few decades later, in 1865, the German physicist Rudolf Clausius coined a term for the proportion of energy that’s locked up in futility. He called it “entropy,” after the Greek word for transformation. He then laid out what became known as the second law of thermodynamics: “The entropy of the universe tends to a maximum.”

Physicists of the era erroneously believed that heat was a fluid (called “caloric”). Over the following decades, they realized heat was rather a byproduct of individual molecules bumping around. This shift in perspective allowed the Austrian physicist Ludwig Boltzmann to reframe and sharpen the idea of entropy using probabilities.

Boltzmann distinguished the microscopic properties of molecules, such as their individual locations and velocities, from bulk macroscopic properties of a gas like temperature and pressure…


Google DeepMind, Google’s flagship AI research lab, wants to beat OpenAI at the video-generation game — and it might just, at least for a little while.

Multi-Agent Collaboration: The Future of Problem Solving with GenAI

The field of artificial intelligence (AI) has witnessed extraordinary advancements in recent years, ranging from natural language processing breakthroughs to the development of sophisticated robotics. Among these innovations, multi-agent systems (MAS) have emerged as a transformative approach for solving problems that single agents struggle to address. Multi-agent collaboration harnesses the power of interactions between autonomous entities, or “agents,” to achieve shared or individual objectives. In this article, we explore one specific and impactful technique within multi-agent collaboration: role-based collaboration enhanced by prompt engineering. This approach has proven particularly effective in practical applications, such as developing a software application.

Ways to Deal With Hallucinations in LLM

Originally published on Towards AI.

One of the major challenges in using LLMs in business is that LLMs hallucinate. How can you entrust your clients to a chatbot that can go mad and tell them something inappropriate at any moment? Or how can you trust your corporate AI assistant if it makes things up randomly?

That’s a problem, especially given that an LLM can’t be fired or held accountable.

Revolutionary AI Model Deciphers Language of Plants for the First Time

PlantRNA-FM, an AI model trained on RNA data from over 1,100 plants, decodes genetic patterns to advance plant science, improve crops, and tackle global agricultural challenges.

A groundbreaking Artificial Intelligence (AI) model designed to decode the sequences and structural patterns that form the genetic “language” of plants has been launched by a research collaboration.

Named Plant RNA-FM, this innovative model is the first of its kind and was developed by a partnership between plant researchers at the John Innes Centre and computer scientists at the University of Exeter.

DeepMind Vice President sees AI on the brink of a fundamental shift towards autonomous agents

DeepMind’s Vice President of Drastic Research and Gemini co-tech lead Oriol Vinyals describes how artificial intelligence is moving from narrowly focused systems toward autonomous agents, and what challenges remain ahead.

Ad.

According to Vinyals, AI is going through a fundamental transformation away from highly specialized systems and toward autonomous agents. Speaking on a company podcast, he explained that early AI systems like AlphaStar, which focused on playing StarCraft, were just the beginning of this development.