Thursday 14, December 2023

Techኢት ማለዳTech☕️

Title:- Apple has announced a major expansion of its self-repair program

The company will now allow iPhone 15 and M2-powered Mac users to fix their own devices. This comes after the tech company opened up the program to include iPhone 14 models and M2-powered laptops earlier this year.
There’s also a brand-new diagnostic tool to help users to get to the bottom of things. Apple already offered a System Configuration tool that came up when the device was in Diagnostics mode, but this goes several steps further. The web-based diagnostic tool is now available in the US and can directly test for a number of common issues (assuming you have a second device on hand to run it.) It will require putting the hardware you'd like to test into Diagnostics mode and entering its serial number into the site.

Apple will attempt to diagnose the problem and offer up a step-by-step repair process. This system can test for issues with the display, software integration, audio output, camera functionality and Face ID, with more tests on the way. This essentially gives regular users the same access to diagnostic tests as Apple authorized service providers.

As always, Apple’s not doing this out of the kindness of its M2-powered heart. The company has been facing increasing pressure from various world governments to abide by newly-mandated right-to-repair laws. This program potentially heads off any legal troubles, though Apple still has some more work to do to cement its right-to-repair commitment. Still, additional DIY repair options are always welcome, no matter how we got here.

Source:- https://www.engadget.com/apples-self-repair-program-now-includes-the-iphone-15-and-more-m2-powered-macs-162040671.html

በሶፍትዌር ዕክል ምክንያት ከ2 ሚሊየን ለሚልቁ የቴስላ ምርቶች ጥሪ ቀረበ

የኤለን መስክ አንዱ ኩባንያ የሆነው“ቴስላ አውቶሞቲቭ” በሶፍትዌር ችግር ምክንያት ከ2 ሚሊየን የሚልቁ ተሸከርካሪዎቹን ለማዘመን ጥሪ ማቅረቡ ተሰምቷል፡፡
ጥሪ የቀረበው የአሜሪካው ቁጥጥር ባለሥልጣን ለሁለት ዓመታት ፍተሻ ሲያደርግ ቆይቶ የመኪናዎቹ የ”አውቶ ፓይለት” ሥርዓት ችግር እንዳለበት በማሳወቁ ምክንያት መሆኑ ተጠቅሷል፡፡

መኪናዎቹ የተገጠመላቸው የ“ድራይቨር አሲስታንስ” ቴክኖሎጂ ሥርዓት (“አውቶፓይለት”) ሾፌሩን የሚያግዝ ክፍል በከፊል ጉድለት እንዳለበት የባለሥልጣኑ የምርመራ ውጤት አረጋግጧል።
በመሆኑም ከፈረንጆቹ 2015 ጀምሮ በአሜሪካ የሚንቀሳቀሱ የ“አውቶ ፓይለት ሥርዓት” የተካተተላቸው ከ2 ሚሊየን የሚልቁ መኪናዎች የሶፍትዌር ዝመና (update) እንዲደረግላቸው ኩባንያው ጥሪ ማቅረቡን ቢቢሲ ዘግቧል፡፡
ኩባንያው ጉዳዩ ቀላል እንደሆነና የመኪናዎቹን ሶፍትዌር ባሉበት በማዘመን ብቻ በአጭር ጊዜ እንደሚፈታ አስታውቋል፡፡
ደንበኞቹ ሶፍትዌራቸውን ለማዘመን ወይም “አፕዴት” ለማድረግ ወደ ኩባንያው ወይም ወደ ቴስላ የሽያጭ ማዕከላት መሄድ እንደማያስፈልጋቸው ባሉበት ሶፍትዌሩን “አፕዴት” በማድረግ ብቻ ችግራቸው እንደሚፈታም አስታውቋል፡፡

ምንጭ: Fbc

December 14, 2023 Addis Ababa

Techኢት NightByte🌙

📰 Title: Meta's AI Empowers Ray-Ban Smart Glasses with Object Recognition and Language Translation Capabilities! 🕶️🌐

Meta has made an exciting announcement regarding the Meta Ray-Ban smart glasses, revealing advanced AI features that will profoundly enhance user experiences. In an upcoming early access test, users will have the opportunity to immerse themselves in the cutting-edge capabilities of these futuristic spectacles.

The integration of Meta's AI technology allows wearers to activate a virtual assistant by simply uttering the phrase "Hey Meta." This assistant possesses the remarkable ability to perceive and understand the surrounding environment, thanks to the glasses' built-in camera and microphones.

One of the most intriguing aspects of this development is the glasses' object recognition capability. The AI assistant can identify various objects in real-time, providing users with valuable information about their surroundings. Additionally, the glasses can translate languages on the fly, opening up a world of possibilities for seamless communication across linguistic barriers.

During this early access phase, Meta aims to refine the glasses' AI functionality by collecting user feedback and addressing any bugs or limitations. This collaborative effort will contribute to the evolution and improvement of the technology.

The potential applications of Meta's AI-powered Ray-Ban smart glasses are vast and captivating. From assisting visually impaired individuals to enhancing productivity in various industries, this technology holds promise for transforming how we interact with the world around us.

Meta's announcement marks a significant step forward in the realm of augmented reality, demonstrating the company's commitment to pushing the boundaries of innovation. As the early access test progresses, users will have the opportunity to be at the forefront of this exciting technological revolution.

Source: https://www.cnet.com/tech/computing/metas-ray-ban-glasses-added-ai-that-can-see-what-youre-seeing/

#MetaAI #RayBanSmartGlasses #AugmentedReality #AIInnovation #TechNews

15, December 2023 Addis Ababa

Techኢት ማለዳTech☕️

Title: Intel Unveils Gaudi3 Chip to Challenge Nvidia and AMD in AI Hardware Market

Introducing Gaudi3 for Generative AI

Intel takes on industry giants Nvidia and AMD with the unveiling of Gaudi3, a powerful chip designed for generative AI applications. Launching next year, Gaudi3 aims to compete head-to-head with Nvidia's H100 and AMD's upcoming MI300X, signaling Intel's strategic entry into the booming AI hardware market.

Core Ultra Chips for Enhanced AI in Windows Devices

At a launch event in New York, Intel CEO Pat Gelsinger underscores the growing role of AI in personal computing. The announcement includes Core Ultra chips tailored for Windows laptops and PCs, integrating NPUs to boost AI performance. While not on the scale of cloud-based models like ChatGPT, these chips showcase Intel's commitment to efficiency, featuring a 7-nanometer process for improved power efficiency.

Xeon Server Chips and Industry Trends

In addition to Core Ultra, Intel introduces fifth-generation Xeon server chips for large organizations and cloud companies. Paired with Nvidia GPUs, these processors excel in AI model deployment, particularly inferencing. This strategic move aligns with the industry trend of meeting the escalating demand for AI chips. Intel's shares see a 1% increase following the announcement, marking a significant step in challenging Nvidia's dominance in the AI hardware market.

Source: https://www.cnbc.com/2023/12/14/intel-unveils-gaudi3-ai-chip-to-compete-with-nvidia-and-amd.html

A 21 hour tutorial of a beginner frontend tutorial. It was given by this guy called Zach Gall who is a full-stack developer. I enjoyed the detailed explanations, the challenges and him going step by step to solve the problems. It focused on Javanoscript, HTML and CSS.

Here are three main things that I learned from watching this 21 hour tutorial.
1. Programmers aren’t necessarily required to remember every way of how things work. ⚙️⚙️

I learnt being a programmer is being comfortable of not knowing everything in mind but being curious enough to google it, understand and apply it when its needed.

Zach specially in the coding challenges when there came some problems he couldn’t solve on the first try just googled it, or read it from the documenation and implemented it.

Which is the next thing I learned:-
2. The Documentation 📑
Everything- every programming language has its documentation on websites like MDN. Reading it I have found very helpful specially with the cases where you have to go deep into a topic in that language. I liked the MDN site for its clear examples.
Learning these days can’t be easier. The resources are outhere. The people giving them are also nice. For example Zach is a likeable guy he is smart and capable but he is not boastful in any way infact he is comfortable to share when he is uncertain on some stuff. The sites i found on the video were great too.

3. There are some helpful sites. 🎁🎁

I found some cool sites that helped me and will help me in the future in this journey.
codpen.io:- has css, html, and JS and shows the effect of your code in real-time
codewars.io:- has a bunch of coding challenges. there are people who solve it on youtube.
frontendmentor.io:- has css challenges ranging from newbie to advanced
github.io:- this site lets you have a server to your web-site. Every one has it. I don’t fully understand how it works but I just understood I needed it.
mastgames.io:- got to practice their free game on flex-box. I liked it a lot. Helped cement flex-box because of the repition and story added to it. (I was directed to this site from codpen.io)
And the following I got to see from random places:-
stackblitz.io:- its literally like codepen with some unique features. It works like a vs-code as it opens files and stuff like such. Going live is possible as well. So you can watch your work in different tab as you are working on it.
leetcode.com:- saw a couple of programmers on youtube talking about it. Apparently its the famous version of codewars. Will try out the challenges in the future.

#coding
#thejourney

December 15, 2023 Addis Ababa
Techኢት NightByte🌙

Title: OpenAI Resolves Major Outage on ChatGPT AI Chatbot

ChatGPT Faces Intermittent Unavailability and Elevated Error Rates; OpenAI Investigating Causes


OpenAI, the leading artificial intelligence research laboratory, announced today the successful resolution of a major outage that occurred on its popular AI chatbot, ChatGPT. The service experienced intermittent unavailability for approximately 40 minutes before normal functionality was restored.

During the outage, users encountered difficulties accessing ChatGPT, with service interruptions reported throughout the duration. OpenAI promptly addressed the issue and worked diligently to rectify the situation, ensuring a swift resolution.

In addition to the intermittent unavailability experienced by ChatGPT, OpenAI acknowledged that some users of ChatGPT Enterprise, the enterprise-grade version of the chatbot tailored for businesses, encountered elevated error rates. The company is actively investigating the causes behind these issues to prevent future disruptions and enhance the overall reliability of the platform.

This recent incident follows a previous disruption earlier this month when approximately 10% of users were unable to send messages to ChatGPT. Furthermore, ChatGPT suffered a major outage in November, highlighting the challenges faced by OpenAI in maintaining uninterrupted service for its rapidly growing user base.

OpenAI has not provided specific details regarding the causes of the recent outages, leaving users and industry experts speculating about potential underlying factors. The company, known for its groundbreaking natural language processing technology, has not released an official statement shedding light on the root causes of these disruptions.

ChatGPT, the AI-powered chatbot developed by OpenAI, has achieved remarkable success, setting records as the fastest-growing consumer application in history. Presently, the platform boasts an impressive user base of approximately 100 million weekly active users. Furthermore, over 92% of Fortune 500 companies have adopted ChatGPT for various business applications, solidifying its position as a leading AI tool in the corporate landscape.

OpenAI has faced recent organizational challenges, including the termination and subsequent reinstatement of CEO Sam Altman in November. Following pressure from employees and investors, Altman was reinstated after a brief period of uncertainty. These internal dynamics, coupled with the recurring outages experienced by ChatGPT, have posed significant tests for OpenAI's leadership and operational stability.

With the resolution of the latest outage, OpenAI is working diligently to regain user confidence and ensure the reliability of ChatGPT moving forward. The company remains committed to addressing the challenges faced by its AI chatbot and improving its services to meet the evolving needs of its vast user base.

Source: https: CNBC

Tesla unveils Optimus Gen 2: its next generation humanoid robot

Sat 16, December 2023

Techኢት ማለዳTech☕️

Title:-OpenAI partners with global news publisher Axel Springer to offer news in ChatGPT

News publisher Axel Springer has partnered with OpenAI, the owner of ChatGPT, to provide AI-generated summaries of news articles. The summaries, including material from subnoscription-based stories, will be available on ChatGPT as soon as the articles are published. OpenAI and Axel Springer, the global news publisher, have struck a deal that allows ChatGPT to summarise news content from outlets under the Axel Springer media brand.

What does this mean for users? ChatGPT’s users could simply inquire about a specific topic, and the AI-powered chatbot would respond with concise summaries from articles sourced from media giants portfolios like Politico, Business Insider, Bild, and Welt. This feature even extends to articles typically accessible only to subscribers. The provided responses will be accompanied by proper attribution and direct links to the full articles, ensuring transparency in information sourcing.

Furthermore, as part of the agreement, Axel Springer will provide content from its media brands to be utilised as training data for OpenAI’s large language models. In return, Axel Springer will receive financial compensation for providing access to its content for training purposes.

The head of intellectual property and content at OpenAI, Tom Rubin, stated that the Axel Springer content will be available to ChatGPT users in the first quarter of 2024. The deal is expected to give Axel Springer content a ‘favourable position’ in ChatGPT search results, aiming to drive traffic and subnoscription revenue to the publisher’s brands.

The deal’s financial terms, which are non-exclusive and span multiple years, were not disclosed. Axel Springer CEO Mathias Doepfner expressed the company’s intentions to explore the opportunities of AI-empowered journalism and elevate the quality, relevance, and business model of journalism to the next level.

Why does it matter?
The news comes as publishers, artists, and writers increasingly weigh or pursue legal action against companies behind popular generative AI tools for allegedly using their content or creations as training data. However, the Axel Springer deal follows OpenAI’s previous agreement with the Associated Press, underscoring the growing intersection between AI technology and news publishers. The deal represents a noteworthy milestone in the evolving relationship between journalism companies and AI firms, and transcends the conventional practice of merely supplying data for training ChatGPT’s models. Instead, it introduces a novel dimension by leveraging vetted journalism to enhance the precision and reliability of ChatGPT’s responses.
Source:-https://dig.watch/updates/openai-partners-with-global-news-publisher-axel-springer-to-offer-news-in-chatgpt

WHO IS WHO in computer science
Bill Gates

Bill Gates (born October 28, 1955) is an American business magnate, philanthropist, investor, computer programmer, and inventor. Gates is the former chief executive and chairman of Microsoft, the world’s largest personal-computer software company, which he co-founded with Paul Allen.

#whoiswho

Revolutionizing the Modeling Landscape: The Rise of AI Model Aitana

In a groundbreaking move, Spanish agency The Clueless has harnessed the power of artificial intelligence to create the first-ever Spanish AI model, Aitana. Born out of a need to address project setbacks tied to real influencers and models, Aitana, a virtual 25-year-old from Barcelona, has become a trailblazer in the modeling industry. Developed during a challenging period for the agency, Aitana's creation aimed to provide a dependable and controllable model for brands, allowing The Clueless to navigate the unpredictability associated with real individuals in the industry. The success of Aitana, earning up to €10,000 monthly and gaining over 121,000 followers on Instagram within a few months, has propelled The Clueless to expand their innovative approach by introducing a second virtual model, Maia, signaling a potential shift in the dynamics of the modeling world.*

*This initiative has sparked interest from various brands eager to embrace personalized virtual models, citing the potential for cost savings and increased flexibility. The Clueless envisions a future where AI-generated models democratize the industry, providing an alternative to traditional influencer collaborations and potentially benefiting smaller companies with limited advertising budgets. While the concept faces criticism for potential societal impacts, including concerns about perpetuating unrealistic beauty standards and overtly sexualized imagery, The Clueless remains steadfast in their belief that this novel approach could bring positive change to an industry often characterized by unpredictability and high costs.

Her instagram account: fit_aitana

WHO IS WHO in computer science

Alfred Aho (born August 9, 1941) is a Canadian computer scientist best known for his work on programming languages, compilers, and related algorithms, and his textbooks on the art and science of computer programming. Aho received a B.A.Sc. in Engineering Physics from the University of Toronto.


🚀🚀 https://news.1rj.ru/str/TechInEthio 🚀🚀


#whoiswho

December 16, 2023 Addis Ababa

Techኢት NightByte🌙

Title:Google's Gemini is the new AI model really better than ChatGPT?

Google Deepmind has recently announced Gemini, its new AI model to compete with OpenAI's ChatGPT. While both models are examples of "generative AI", which learn to find patterns of input training information to generate new data (pictures, words or other media), ChatGPT is a large language model (LLM) which focuses on producing text. In the same way that ChatGPT is a web app for conversations that is based on the neural network know as GPT (trained on huge amounts of text), Google has a conversational web app called Bard which was based on a model called LaMDA (trained on dialogue). But Google is now upgrading that based on Gemini.

What distinguishes Gemini from earlier generative AI models such as LaMDA is that it’s a “multi-modal model”. This means that it works directly with multiple modes of input and output: as well as supporting text input and output, it supports images, audio and video. Accordingly, a new acronym is emerging: LMM (large multimodal model), not to be confused with LLM.

In September, OpenAI announced a model called GPT-4Vision that can work with images, audio and text as well. However, it is not a fully multimodal model in the way that Gemini promises to be

For example, while ChatGPT-4, which is powered by GPT-4V, can work with audio inputs and generate speech outputs, OpenAI has confirmed that this is done by converting speech to text on input using another deep learning model called Whisper. ChatGPT-4 also converts text to speech on output using a different model, meaning that GPT-4V itself is working purely with text.

Likewise, ChatGPT-4 can produce images, but it does so by generating text prompts that are passed to a separate deep learning model called Dall-E 2, which converts text denoscriptions into images.

In contrast, Google designed Gemini to be “natively multimodal”. This means that the core model directly handles a range of input types (audio, images, video and text) and can directly output them too. The distinction between these two approaches might seem academic, but it’s important. The general conclusion from Google’s technical report and other qualitative tests to date is that the current publicly available version of Gemini, called Gemini 1.0 Pro, is not generally as good as GPT-4, and is more similar in its capabilities to GPT 3.5.


🚀🚀 https://news.1rj.ru/str/TechInEthio 🚀🚀

https://theconversation.com/googles-gemini-is-the-new-ai-model-really-better-than-chatgpt-219526#:~:text=The%20verdict&text=The%20general%20conclusion%20from%20Google's,its%20capabilities%20to%20GPT%203.5.