CybAI news - - Page 2985

30 May

The ugly truth behind ChatGPT: AI is guzzling resources at planet-eating rates | Mariana Mazzucato

stp2y0 CommentsChat-GPT

When you picture the tech industry, you probably think of things that don’t exist in physical space, such as the apps and internet browser on your phone. But the infrastructure required to store all this information – the physical datacentres housed in business parks and city outskirts – consume massive amounts of energy. Despite its name, the infrastructure used by the “cloud” accounts for more global greenhouse emissions than commercial flights. In 2018, for instance, the 5bn YouTube hits for the viral song Despacito used the same amount of energy it would take to heat 40,000 US homes annually.This is…

30 May

Robot, can you say ‘cheese’?

stp2y0 CommentsGenAIRobotics Research; Engineering; Engineering and Construction; Vehicles; Robotics; Artificial Intelligence; Neural Interfaces; Computer Modeling

What would you do if you walked up to a robot with a human-like head and it smiled at you first? You'd likely smile back and perhaps feel the two of you were genuinely interacting. But how does a robot know how to do this? Or a better question, how does it know to get you to smile back? Source link lol

30 May

India Needs a Broad Range of Suppliers to Succeed in Chips

stp2y0 CommentsRAG models

India will have to attract a wider range of semiconductor suppliers to succeed in building a robust chip industry in the country, according to the leader of an influential trade association. Source link lol

30 May

Ultra fast joins with multi-join | Deephaven

stp2y0 CommentsNews

Deephaven's version 0.28 release brought some cool new features and improvements to client libraries. Perhaps the most exciting is the addition of multi_join to the Deephaven table API. With multi_join, you can accomplish what previously took several lines of code in a single line. Not only that, but it's way faster than the old way of chaining joins together.Multi-join is a new feature that joins three or more tables together. It's faster, much more memory efficient than the only alternative, and works all the same on static and dynamic data.In this blog, we'll compare performance of using multi_join and its…

30 May

TinyAgent: Function Calling at the Edge

stp2y0 CommentsAIedge deployment, function calling, retrieval-augmented generation, tinyagent

The ability of LLMs to execute commands through plain language (e.g. English) has enabled agentic systems that can complete a user query by orchestrating the right set of tools (e.g. ToolFormer, Gorilla). This, along with the recent multi-modal efforts such as the GPT-4o or Gemini-1.5 model, has expanded the realm of possibilities with AI agents. While this is quite exciting, the large model size and computational requirements of these models often requires their inference to be performed on the cloud. This can create several challenges for their widespread adoption. First and foremost, uploading data such as video, audio, or text…

30 May

Addressing the EU’s Stricter Standby Power Consumption Standards – EE Times

stp2y0 CommentsNews

//php echo do_shortcode('[responsivevoice_button voice="US English Male" buttontext="Listen to Post"]') ?> Starting in May 2025, most household appliances sold in the EU must have a standby power consumption of less than 300mW, a significant reduction from the current 500mW limit. To help meet these stringent standards, Power Integrations (PI) has released a new application guide (AN-109) for designing small open-frame power supplies. This guide features PI’s LinkSwitch-XT2SR non-isolated flyback switching IC, which boasts up to 90% efficiency and a no-load power consumption of less than 5mW. As IC power consumption inevitably increases to support more powerful functions, the LinkSwitch-XT2SR ensures overall…

30 May

How Retrieval Augmented Generation (RAG) Work

stp2y0 CommentsNewscoding, community, development, engineering, gpt3, inclusive, LLM, machinelearning, rag, software

Retrieval Augmented Generation (RAG, pronounced 'rag') works by fetching selective data from a custom knowledge base and integrating it with the output of a language model to provide accurate and up-to-date responses.RAG can be defined as a ChatGPT-like interface that can use your pdfs, documents or databases to answer questions from you. You can use it as a study assistant to understand documents by asking questions about those documents. In this article, we discuss the benefits of using a RAG system and explain its key components. We also detailhow RAG works to enhance the capabilities of large language models by…

30 May

The US gave sensitive plans for over 1,000 American weapons to Ukraine, says 2 officials who gave only a cryptic hint as to what they are: report

stp2y0 CommentsRAG models

The US has given Ukraine manufacturing plans for more than 1,000 American weapons in hopes of helping Kyiv bolster its own arms production, two officials told The New York Times.The military officials told NYT's John Ismay of the transfer during a reporting visit to a new factory for Howitzer artillery shells near Dallas.According to the NYT, these two officials were William A. LaPlante, Undersecretary of Defense for Acquisition and Sustainment, and Douglas R. Bush, Assistant Secretary of the Army for Acquisition, Logistics, and Technology.They told the outlet that the US has also translated technical manuals from English to Ukrainian but…

30 May

Foreign Payments and AI Tools – Transforming Global Transactions

stp2y0 CommentsViral News

In today's interconnected world, the ability to make and receive payments across borders is essential for businesses and individuals alike. Traditionally, foreign payments have been fraught with challenges, including high fees, long processing times, and complex regulatory requirements. However, the advent of artificial intelligence (AI) is revolutionizing this landscape, making international transactions faster, cheaper, and more secure. The Evolution of Foreign Payments Traditional Methods and Challenges Historically, foreign payments have relied on a network of banks and financial institutions to process transactions. Methods such as wire transfers and correspondent banking systems have been the mainstay. These methods, however, come with…

30 May

FunSearch: Making new discoveries in mathematical sciences using Large Language Models

stp2y0 CommentsNews

Research Published 14 December 2023 Authors Alhussein Fawzi and Bernardino Romera Paredes By searching for “functions” written in computer code, FunSearch made the first discoveries in open problems in mathematical sciences using LLMsLarge Language Models (LLMs) are useful assistants - they excel at combining concepts and can read, write and code to help people solve problems. But could they discover entirely new knowledge?As LLMs have been shown to “hallucinate” factually incorrect information, using them to make verifiably correct discoveries is a challenge. But what if we could harness the creativity of LLMs by identifying and building upon only their very…