Good Morning! In today’s news we have a new supercomputer DeepSouth mimics human brain to revolutionize AI. Google settles Play Store lawsuit for $700 million, agrees to reforms allowing more competition. Ruby's updated just-in-time compiler YJIT boosts performance in Ruby 3.3, but monitors memory usage.

— Forrest Knight & Meghanadh Vasireddy

New Supercomputer MIMICS the Human Brain

Australian scientists are set to launch a supercomputer that can simulate the human brain's capacity. Named DeepSouth, this supercomputer is being built by the International Centre for Neuromorphic Systems (ICNS) in Sydney, Australia, in collaboration with two of the world's leading computer technology manufacturers.

DeepSouth is designed to perform about 228 trillion operations per second, similar to the estimated operations of a human brain. Unlike traditional computers, DeepSouth will use interconnected artificial neurons and synapses to complete tasks, essentially replicating the brain's capacity for learning and adaptation.

This supercomputer belongs to an approach known as neuromorphic computing, which aims to mimic the biological processes of the human brain. The goal is to understand how our brains process massive amounts of information while consuming minimal power. Our brains can perform the same number of operations per second with just 20 watts of power, while weighing just 1.3kg-1.4kg. In contrast, supercomputers require large amounts of electrical power to run.

DeepSouth is expected to be operational by 2024 and will be run from the International Centre for Neuromorphic Systems at Western Sydney University. This development could revolutionize our understanding of how our brains work and could lead to the creation of more advanced and efficient AI systems.

Read More Here

Google settles Play Store antitrust lawsuit

Google has agreed to a $700 million settlement to resolve an antitrust lawsuit over its Play Store practices, a case brought by all 50 U.S. states. The tech giant will pay $630 million in restitution to consumers and an additional $70 million to the states for their sovereign claims.

The lawsuit, initiated in 2021, accused Google of maintaining an illegal monopoly that stifled competition and harmed consumers through high commissions and restrictive contracts. The settlement, disclosed in a recent filing, requires Google to reform its business practices to foster greater competition in the app market.

Under the terms of the settlement, Google will make several changes to its Play Store policies. These include allowing developers to use their own payment systems without being penalized and providing more transparency in its app promotion processes. The settlement also mandates that Google must not impose any restrictions on developers launching their apps on other app stores at the same time as on the Play Store.

The agreement comes after a federal jury found Google's practices around the Play Store and its billing system to be anticompetitive. This verdict was part of a separate but related lawsuit filed by Epic Games, which Google plans to appeal.

While the settlement has been hailed as a significant win for consumers and a step towards more open competition, some, like Epic Games CEO Tim Sweeney, argue that it does not go far enough in addressing the core issues of app store monopolies.

The settlement is pending approval by a judge, with a hearing scheduled for February 2024. If approved, consumers who made purchases on the Google Play Store between August 2016 and September 2023 may be eligible for restitution as part of the $630 million allocated for consumer claims.

Read More Here

Ruby 3.3's YJIT: A Leap Forward in Performance and Efficiency

The latest version of Ruby's Just-In-Time (JIT) compiler, YJIT, included in Ruby 3.3, is making waves in the developer community with its significant performance improvements and reduced memory usage.

YJIT, developed by the Ruby team, has been optimized to provide better performance across the board, warming up faster and using less memory. Notably, rendering liquid templates is now over 2.5x faster with YJIT enabled than it is using only the Ruby interpreter. Railsbench, a benchmarking tool for Ruby on Rails applications, is about 65% faster with YJIT.

In addition to performance enhancements, YJIT 3.3 introduces new features to simplify developers' lives. One such feature is the ability to enable YJIT at runtime from Ruby code by calling RubyVM::YJIT.enable. This allows developers to selectively enable YJIT on some forked processes without enabling it on all processes, and to enable YJIT only after the application has finished booting.

Despite the improvements, it's important to note that YJIT will use more memory than the Ruby interpreter because the JIT compiler needs to generate machine code in memory and maintain additional state information. However, the memory usage for metadata of compiled code has been significantly reduced in Ruby 3.3, making YJIT more memory-efficient.

The Ruby community has responded positively to these improvements. Shopify reported a 15% speed increase in their production code with YJIT 3.3. However, some users have reported increased memory usage with YJIT enabled. It's recommended to monitor memory usage and adjust the --yjit-exec-mem-size option as needed.

Read More Here

Apple Develops Breakthrough Method for Running LLMs on iPhones

Apple researchers have developed an innovative technique that allows large language models (LLMs) to run on iPhones, overcoming typical memory constraints associated with such deployments on devices with limited resources. LLMs, like ChatGPT and Claude, are known for their data and memory intensity, demanding vast memory resources to function. Apple's method utilizes flash memory, the same storage used for apps and photos, to store AI model data. This breakthrough enables AI models to operate up to twice the iPhone's available memory, resulting in a significant boost in speed on CPUs and GPUs.

This advancement opens doors for future iPhone capabilities, including enhanced Siri features, real-time language translation, and more sophisticated AI applications in photography and augmented reality. It also paves the way for on-device execution of complex AI assistants and chatbots, reportedly an area of Apple's ongoing development.

Read More Here

Youtube Spotlight

iPhone Thief Explains How He Breaks Into Your Phone | WSJ

The WSJ interviewed Aaron Johnson, a thief who exploited a vulnerability in Apple’s software to steal and profit off hundreds of iPhones, leading to a prison sentence of 94 months. He targeted people in bars to obtain their phone passcodes and then used them to access bank accounts, change Apple ID passwords, and sell the phones for profit. Apple has since implemented new security measures to address these vulnerabilities.

Was this forwarded to you? Sign Up Here