Good morning! This is your daily ☕️ Techpresso.
In today's Techpresso:
🍎 Apple poached 30+ Google experts to open a secret AI lab
👽 Mysterious “gpt2-chatbot” AI model appears suddenly, confuses experts
🤖 Amazon releases Q business chatbot
🌐 Apple reportedly working on AI-enabled Safari
⚖️ Binance founder sentenced to four months for money laundering
🐰 Hyped Rabbit R1 AI device faces harsh criticism in first reviews
📰 8 major newspapers join legal backlash against OpenAI, Microsoft
🎁 + 10 other news you might like
🔮 + 6 handpicked research papers and tools
🍎 Apple poached 30+ Google experts to open a secret AI labLINK
Apple has reportedly opened a secret AI research lab in Zurich, known as the 'Vision Lab,' after hiring at least 36 AI experts from Google. The Zurich-based 'Vision Lab,' led by former Google AI head John Giannandrea, has already produced significant research in generative AI, focusing on models that interpret text and imagery to deliver precise results. Despite Apple's silent approach in AI research, leading to perceptions of its lateness in the AI race, the company has been discreetly advancing cutting-edge AI technology and maintaining a low profile in recruitment and product development.
👽 Mysterious “gpt2-chatbot” AI model appears suddenly, confuses expertsLINK
A new mysterious chatbot named "gpt2-chatbot" appeared in the LMSYS Chatbot Arena, sparking speculation it might be a test version of OpenAI's upcoming GPT-4.5 or GPT-5.
The gpt2-chatbot model is restricted to eight queries per day in the site's arena mode, limiting extensive testing and evaluation by users.
Despite initial excitement and speculations about its capabilities surpassing current models, user tests and evaluations suggest that the gpt2-chatbot does not represent a significant leap over GPT-4.
🤖 Amazon releases Q business chatbotLINK
Amazon is enhancing its enterprise AI offering by making its AI chatbot Q more widely available to AWS users as an assistant for various tasks including coding and HR inquiries. The chatbot Q is part of Amazon's larger strategy to focus on business-to-business AI services through AWS, with it supporting companies by learning from their data and workflows, and offering features like coding assistance and app generation. Amazon has introduced new functionalities to Q such as Amazon Q Developer and Amazon Q Apps, designed to aid in coding, app testing, and building generative AI-based apps without requiring users to have coding expertise.
🌐 Apple reportedly working on AI-enabled SafariLINK
Apple is developing Safari 18 with features like a personalized user interface, "Web Eraser," and AI-powered "Intelligent Search," set to release alongside iOS 18 and macOS 15.
"Intelligent Search" in Safari 18 will use Apple's Ajax language-learning model for on-device AI technology to summarize webpage content.
Safari 18's "Web Eraser" feature will allow users to delete specific parts of web pages, enhancing privacy by remembering these changes even after closing the tab or window.
⚖️ Binance founder sentenced to four months for money launderingLINK
Changpeng Zhao, founder of Binance, was sentenced to four months in prison for violating US anti-money laundering laws, despite prosecutors' request for a three-year sentence.
Zhao pleaded guilty to money laundering charges and stepped down from Binance, agreeing to a settlement that includes paying $4.3 billion in fines and a personal fine of $50 million.
The settlement acknowledges Binance's evasion of anti-money laundering regulations, allowing transactions from countries under US sanctions, and includes agreements with the CFTC and SEC.
🐰 Hyped Rabbit R1 AI device faces harsh criticism in first reviewsLINK
The Rabbit R1, an AI-powered assistant device launched amidst high anticipation, has received predominantly negative reviews questioning its necessity and comparing it unfavorably with simpler app-based solutions.
Despite its innovative technology and strong pre-order sales, critics argue the Rabbit R1 is underdeveloped, with limited app compatibility, poor battery life, and missing features, suggesting it was rushed to market.
Attempts by reviewers to replicate the Rabbit R1's functionality on a smartphone have sparked debate, with Rabbit's CEO defending the device's bespoke operating system and cloud-based AI model against claims it could be simply an app.
📰 8 major newspapers join legal backlash against OpenAI, MicrosoftLINK
News organizations including the New York Daily News, Chicago Tribune, and five others, represented by Alden Global Capital, have sued OpenAI and Microsoft for alleged copyright infringement, claiming unauthorized training on their articles.
The plaintiffs provided evidence showing ChatGPT and Copilot producing excerpts of articles without permission, as well as Copilot fetching complete articles verbatim from the web shortly after publication without proper attribution.
The lawsuit argues against the use of copyrighted newspaper content by tech companies to enrich their AI technologies without compensation, amidst broader legal actions by other news entities against OpenAI and Microsoft.
Other news you might like
Microsoft feared falling behind Google's AI advances, internal emails reveal.LINK
ChatGPT now has a ‘memory’.LINK
Google Chrome is getting a smarter, savvier address bar thanks to AI.LINK
Amazon sales soar with boost from artificial intelligence and advertising.LINK
Microsoft and Axel Springer plan ad-funded AI chatbots for news.LINK
Boston Dynamics unveils its fur-covered dancing robot dog 'Sparkles'.LINK
Google fired a software engineer over an anti-war demonstration — he says he wasn’t even protesting.LINK
Reddit users compile list of words and phrases that unmask ChatGPT's writing style.LINK
Bill comes due for money-losing tech companies that borrowed billions.LINK
AWS S3 storage bucket with unlucky name nearly cost developer $1,300.LINK
Latest research and tools
Better and Faster Large Language Models via Multi-Token Prediction: this paper introduces an approach for improving the efficiency and performance of large language models through the prediction of multiple tokens simultaneously.LINK
Kolmogorov-Arnold Networks: a promising alternative to Multi-Layer Perceptrons with activation functions on edges instead of nodes, offering better accuracy, interpretability, and potential for scientific discovery.LINK
Neofetch: a command-line tool that displays system information alongside your operating system's logo in a visually pleasing way, with customization options for the displayed content and almost 150 operating system support.LINK
Building a Large Japanese Web Corpus for Large Language Models: the paper discusses the methodology for creating a substantial Japanese web corpus to train large language models.LINK
Replacing Judges with Juries: Evaluating LLM Generations with a Panel of Models: the paper discusses an approach to assess large language models (LLMs) through collective evaluation by multiple model-generated panels instead of a single expert judgment.LINK
Iterative Reasoning Preference Optimization: the paper introduces a method for enhancing machine learning models by incrementally incorporating user preferences to improve decision-making accuracy.LINK
Want to get the latest news differently? Find us on:
See you tomorrow for a new dose of ☕️ Techpresso!
-min.png){kind=small}
.png){kind=small}
