- NATURAL 20
- Posts
- MOBILE ALOHA Google's new OPEN SOURCE robot
MOBILE ALOHA Google's new OPEN SOURCE robot
PLUS: What to expect at CES 2024, Apple's New Gen AI-Powered Siri and more
Today:
MOBILE ALOHA Google's new OPEN SOURCE robot
Today's spotlight shines on a revolutionary development in the world of robotics that's easy to miss amidst the flurry of tech news but stands out for its potential impact on everyday life. Dubbed "Mobile Aloha," this innovation is a game-changer in making advanced robotics accessible and customizable for personal and business use. Here's a rundown of what makes Mobile Aloha the buzz of the town.
Mobile Aloha is an open-source project that's changing the game in robotics. Priced around $32,000, this isn't just another high-brow tech toy. It's an accessible, customizable robotic system you can have in your home, office, or garage.
What to expect at CES 2024: AI, AI and more AI
CES 2024 in Vegas is all about AI. We're talking about a ChatGPT buddy robot, a smart belt guiding the visually impaired, and AI-powered house stuff. This tech fest is the place to be for big reveals, like last year's self-driving stroller. It's also a hot spot for business deals and spotting tech trends. Expect to see everything AI, from Intel's chips to Samsung's smart fridges. There’s talk about AI's impact on jobs and ethics too.
About 130,000 folks will be there, with 4,000 exhibitors and talks from top execs. And while Apple's skipping it, companies from all sectors, even beauty, are jumping in. Besides AI, look out for cool stuff like foldable displays and high-tech wearables. Some companies might upgrade older tech with AI, making things like voice-controlled gadgets smarter. CES is a tech trend thermometer, but with so much going on, it's tough for one thing to stand out - unless it's AI.
Apple May Launch A New Gen AI-Powered Version Of Siri At WWDC 2024: What To Expect
Alright, check this out: Apple's got something big cooking for Siri at their 2024 WWDC. Word on the street is, they're pumping a cool billion into making Siri way smarter with some next-level AI tech. Think smoother chats, like you're talking to a real person. Plus, Siri might be breaking out of the Apple bubble, playing nice with other apps and services. No official word yet, but June's looking like showtime for all the juicy details. This could totally change the game for how we all use Apple stuff.
Dell's New XPS Lineup: Futuristic Design, with Built-in AI
Dell's shaking things up with its new XPS lineup, introducing the XPS 16 and 14, and updating the XPS 13. These sleek machines are all about minimalistic design, using sturdy materials like CNC machined aluminum and Gorilla Glass, in cool colors like Graphite and Platinum.
But the big news? AI-enabled features powered by Intel Core Ultra processors and Windows 11. This tech boosts creativity, performance, and battery life. Speaking of battery, these laptops can charge up to 80% in just an hour.
Isomorphic inks deals with Eli Lilly and Novartis for drug discovery
Isomorphic Labs, a London-based company born from Google's AI division, just landed a huge deal with drug bigwigs Eli Lilly and Novartis. They're teaming up to use AI for drug discovery, a deal worth about $3 billion. Isomorphic gets $45 million from Eli Lilly, with a chance for $1.7 billion more if things go well, and Novartis is throwing in $37.5 million upfront, plus up to $1.2 billion later.
They're using DeepMind’s AlphaFold 2 tech to figure out human protein structures, hoping to find new ways to fight diseases. While the tech isn’t bulletproof, it's made strides, like in liver cancer and neglected diseases. Isomorphic, which started in 2021, is feeling the heat to turn a profit after a $3 million loss in its first year.
Hatz AI Raises $2.5M to Enable MSPs to Deliver AI-as-a-Service
Hatz AI, co-founded by Jimmy Hatzell and Aidan Kehoe, just launched a platform for Managed Service Providers (MSPs) to offer AI services to their clients. They've raised $2.5 million to kick things off. This platform will let MSPs provide AI apps, agents, and other techie stuff, making it easier for businesses to get into AI.
Jimmy Hatzell says MSPs will be big players in spreading AI use. The first products, including an AI app builder, will be out in March. Big names like Matt Higgins from "Shark Tank" are backing them. Hatz AI is all about helping MSPs create and manage AI services easily.
🧠RESEARCH
SIGNeRF, by Dihlmann and team at the University of Tübingen, introduces a quick, controllable way to tweak 3D scenes using Neural Radiance Fields (NeRF). It cleverly combines image diffusion models with NeRF, allowing for editing within photorealistic scenes without the usual complexities. This approach generates 3D consistent images efficiently, using a set of altered images as a reference. The innovation lies in its ability to update scenes in one shot and precisely control edits through depth conditioning.
The Ohio State University team explores GPT-4V(ision), a large multimodal model, for web tasks. Their project, SEEACT, uses GPT-4V for web-based tasks, integrating visual understanding and action. Tested on the MIND2WEB benchmark and live websites, it outperforms text-only models, completing 50% of tasks with manual action grounding. However, grounding remains challenging, indicating potential for further improvement.
"Instruct-Imagen" is a new model by Google's teams that excels in creating diverse images from complex instructions. It combines natural language with various modes like text, style, and subject, making image generation more precise. It's trained in two stages to understand and generate images based on multi-modal inputs. This model outperforms previous ones and shows great potential in handling new, intricate tasks.
🛠️TOP TOOLS
Monica - all-in-one AI assistant equipped with the most advanced AI models (GPT-4, Claude, Bard, etc.) to help you Chat, Search, Write, Translate and more. It also offers tools for image, video, and PDF processing.
Hoppy Copy - generate powerful content for hundreds of different email marketing campaigns, drips, newsletters and more—in seconds
Predis.ai - social media tool that blends features of ChatGPT, Canva, and Hootsuite. It helps in creating and sharing AI-generated videos, carousels, and image posts in brand-specific language.
Flick - marketing platform designed to enhance social media content creation and management. It streamlines brainstorming, writing, scheduling, and optimizing processes.
Apollo.io - sales intelligence and engagement platform. Features include AI-driven lead scoring, data enrichment, and sales engagement tools like email sequences and call automation.
Huge news for @OpenAIDevs: API key based usage is here 🥳
To get started, head to the API key page and generate a tracking token for each key which will enable per key tracking in the usage dashboard for all new requests!
— Logan.GPT (@OfficialLoganK)
10:36 PM • Jan 5, 2024
🗞️MORE NEWS
Wizards Of The Coast, ‘Apex Legends’ Under Fire For AI Art
"Wizards of the Coast and 'Apex Legends' face heat for using AI art. Wizards, accused of using AI for Magic: The Gathering promo, denied it but evidence suggests otherwise. Artist Dave Rapoza cut ties over their stance. Similarly, 'Apex Legends' promotions show AI traces. Both cases highlight the growing, controversial use of AI in art, raising issues about originality and the future of human artists in the industry. Wizards later admitted to AI use, citing Adobe Photoshop's AI tools." FORBES
Apple’s Biggest Challenges in 2024 Have Little to Do With the iPhone
In 2024, Apple's biggest challenges aren't about the iPhone but involve catching up in AI, adapting to App Store changes, and launching the Vision Pro headset. Despite facing a possible fifth quarter of declining sales, Apple is expected to grow iPhone revenue. However, the company lags in AI, with offerings set to debut years after competitors. The Vision Pro, Apple's first new product category in nearly a decade, may not significantly impact revenue soon. App Store regulation changes, particularly in the EU, will force Apple to adapt, potentially affecting profits. Despite these hurdles, Apple's enduring consumer appeal and the strength of its other products, like the iPhone, provide some stability. BLOOMBERG
World's first AI-powered restaurant soon opening doors in Pasadena
Pasadena's soon-to-open CaliExpress is the first AI-powered, fully autonomous restaurant. Created by Miso Robotics and PopID, it features advanced AI and robotic cooking. Customers order at self-service kiosks and can watch robots make fresh burgers and fries. The simple menu offers burgers, wraps, fries, and drinks. CaliExpress aims for safety and efficiency, reducing kitchen risks and waste. It's also a showcase for AI development, inviting schools for tours. CBS NEWS
New research harnesses AI and satellite imagery to reveal the expanding footprint of human activity at sea
A new study in Nature by Global Fishing Watch uses AI and satellites to map the ocean's industrial use, previously hidden from the public. It found 75% of industrial fishing vessels and 25% of transport/energy vessels are untracked, mostly around Africa and South Asia. This groundbreaking research, analyzing 2 million gigabytes of data from 2017-2021, revealed "dark fleets" in marine protected areas and regions previously showing little activity. It also noted a 12% drop in fishing during COVID-19, while offshore energy, especially wind turbines, surged. This work opens new avenues for ocean management and environmental protection. PHYS ORG
AI model predicts future spread of SARS-CoV-2 variants using genomic data
Researchers developed an AI model to predict the spread of new SARS-CoV-2 variants, as detailed in PNAS Nexus. The model analyzed nine million genomic sequences from 30 countries, identifying patterns in large infection waves. It combined genomic data with information on COVID-19 cases, vaccination rates, and non-pharmaceutical interventions. The AI detected 73% of variants causing significant outbreaks within a week of their emergence, improving to 80% with two-weeks' data. The study's findings, with high accuracy rates (AUC values of 86% for one week and 91% for two weeks), demonstrate the model's effectiveness in early detection and risk assessment of new variants. This approach could significantly enhance pandemic preparedness by integrating genetic factors into epidemiological models. NEWS MEDICAL LIFE SCIENCES
Chinese Companies Aren’t Interested In NVIDIA’s Slower AI Chips
Nvidia, a top AI chipmaker, introduced a slower, less powerful AI chip for China, complying with US export rules. However, major Chinese firms like Alibaba, Tencent, Baidu, and ByteDance (TikTok's owner) are showing little interest in these downgraded chips. These companies might reduce their orders from Nvidia, as the performance gap narrows between these chips and Chinese-made alternatives. This shift is partly due to concerns about further US restrictions. The US aims to limit China's technological growth, leading Chinese tech companies to adapt to decreased access to Nvidia's products. Nvidia's CEO, Jensen Huang, likened US export controls to having "our hands tied behind our back." China is a key market for Nvidia, accounting for about 20% of its revenue and dominating China's $7 billion AI chip market with over a 90% share. BUSINESS INSIDER
A New Kind of AI Copy Can Fully Replicate Famous People. The Law Is Powerless.
Martin Seligman, an 81-year-old American psychologist, recently got a surprise from a former student, Yukun Zhao. Zhao's team created a "virtual Seligman" AI that mimics Seligman's thinking and speech, using all his written work. Seligman, intrigued, shared it with friends and family, including his impressed wife. This AI, made in China without Seligman's knowledge, is part of a growing trend of AI chatbots simulating real people. In the U.S., lawmakers are trying to address this with the NO FAKES Act, aiming to let individuals authorize or sue for unauthorized AI replicas. However, this wouldn't affect AI development globally, as seen with AI Seligman in China and AI Perel in the U.S. POLITICO
Copyright law is AI's 2024 battlefield
In 2024, AI's big headache is copyright law. The New York Times is suing OpenAI and Microsoft, accusing them of nicking content. It's a showdown year, with creators already suing to keep their stuff from being snatched by AI. This legal rumble could really shake up AI's future, deciding who gets ahead and who's left behind. Some think the law can handle it, suggesting AI tools just summarize instead of copying stuff outright. But, if the law sides with big companies, smaller players might get squeezed out, unable to afford the data or the legal fights. AXIOS
The AI hardware era is upon us, like it or not
The AI hardware era is rapidly evolving, with major tech companies integrating generative AI into various devices. Here's a roundup of what's happening in 2024:
Google: The Pixel 8 line features photo editing tools using on-device generative AI, like "Magic Editor" for photo manipulation and "Best Face" for optimal facial expressions in photos. The Pixel 8 Pro update added AI summaries for audio recordings and smarter auto-reply suggestions in messaging apps, using a Nano version of Google's Gemini language model. Rumors suggest further AI integration in the Pixel 9 and possible AI-enabled glasses.
Apple: While Apple's approach to generative AI has been quieter, they're rumored to be planning significant AI features. This includes generative AI in Apple Music and iWork, potentially for playlist creation and generating slide decks. Apple is also expected to enhance Siri with large language models in iOS 18 for complex commands. They're reportedly building AI servers and exploring on-device processing for AI applications.
OpenAI: There are no official hardware announcements, but OpenAI has reportedly discussed consumer devices with Jony Ive, possibly backed by a significant investment from SoftBank.
Windows PCs: Intel's 14th-gen Core Ultra chips include a Neural Processing Unit for on-device generative AI tasks, with software optimizations from Adobe and Zoom. AMD's Ryzen 8000 chips are following suit, enhancing AI features in collaboration with Microsoft and software partners.
Android Phones: Qualcomm's Snapdragon 8 Gen 3 processor promises generative AI capabilities, like on-device image generation and language processing. Honor confirmed a virtual assistant for its Magic 6 smartphone using this technology. Samsung is also expected to join the AI trend, possibly with its own large language model, "Gauss," for the Galaxy S24 series.
Miscellaneous Gadgets: Expect to see generative AI in various devices like wearables, smart speakers, and more. An example is Humane's AI pin, a wearable device with ChatGPT capabilities.
The challenge for consumers will be discerning truly useful AI features from mere gimmicks or repackaged existing technologies. The key question will be whether these AI features provide real utility beyond what's achievable with existing cloud-based AI services like ChatGPT.
What'd you think of today's edition? |
What are MOST interested in learning about AI?What stories or resources will be most interesting for you to hear about? |
Reply