This week’s cover image is a Van Gogh mixing board, inspired by MidJourney’s launch of Style Tuner, an incredibly fun tool that ensures consistent styles across AI images.
Executive Summary
- White House regulation is hot off the presses
- AI imagery tool MidJourney launched a cool feature called “tuning” to generate consistent images
- Apple has significant plans for AI in 2024 – don’t count them out
- Google is investing $2 Billion in OpenAI’s competitor, Anthropic. Notably, this is the same company that just got an investment from Amazon.
Top Four Stories
If you only explore four topics this week:
#1 MidJourney Style Tuner for AI Images
Less than 7 hours since MidJourney dropped its new feature Style tuner. People are trying it and seeing a ton of potential. 10 impressive examples and tutorials:
https://twitter.com/_Borriss_/status/1719981444213424552
#2 Apple Is Investing In AI
Kuo: Apple Could Spend $4.75 Billion on AI Servers in 2024
https://www.macrumors.com/2023/10/23/apple-ai-server-spending-2024/
Inside Apple’s Big Plan to Bring Generative AI to All Its Devices
#3 White House AI Regulation
10 new mandates that’ll shake up the industry
https://mashable.com/article/white-house-drops-ai-regulation-bombshell
With Executive Order, White House Tries to Balance A.I.’s Potential and Peril
President Biden announced regulations on Monday that seemed to have a little bit for everyone.
https://www.nytimes.com/2023/10/31/technology/executive-order-artificial-intelligence-regulation.html
TL;DR / key actions across federal agencies:
#4 Google agrees to invest up to $2 billion in OpenAI rival Anthropic
The Rest: AI News of The Week
Don’t let the volume overwhelm you. Have fun and skim it. The links are organized by topic, sorted from ‘coolest’ to ‘least cool’, and each topic is clearly defined with a headline. I do the work so you don’t have to! The links descriptions are often pulled directly from tweets or articles, so it’s not always my voice. Pause when you see something that interests you. Reach out to me any time. I enjoy sharing and discussing these items!
Design/Creativity/Imagery
Artist Showcase
AI artist of the week: @hugovntr with his photorealistic images. 10 of his works:
Shutterstock AI Library
Shutterstock Integrates Creative AI into Library of 700M Images to Offer First-Ever Marketplace of Fully Customizable Stock
Amazon AI Images for Advertisers
Amazon rolls out AI-powered image generation to help advertisers deliver a better ad experience for customers
https://www.aboutamazon.com/news/innovation-at-amazon/amazon-ads-ai-powered-image-generator
A nice introduction to prompting for text-to-images
Prompt tips and examples for image creation
https://www.oneusefulthing.org/p/working-with-ai-two-paths-to-prompting
Stability AI (off my radar, practically speaking, but worth knowing)
Stability AI Previews Enhanced Image Offerings: APIs for Business & New Product Features
https://stability.ai/news/stability-ai-enhanced-image-apis-for-business-features
The Beatles’ final song is now streaming thanks to AI
Machine learning helped Paul McCartney and Ringo Starr turn an old John Lennon demo into what’s likely the band’s last collaborative effort.
https://www.theverge.com/2023/11/2/23943290/now-and-then-the-beatles-new-song-ai
AR/VR/Video
Image to Video, Text to Video
The leading company is Runway, and they just launched a major update
10 Examples of Good Usage of Runway’s Video Update
Incredible Tracking of Soccer Players on a Field
Computer vision can breathe new life into “dumb” sensors like cameras recording a game.
Teams are already exploiting such insights internally to improve player/team performance.
The next step is using this intelligence to transform the viewer experience.
The Apple Event Was Shot on an iPhone
This is how they shot the Apple Event on the iPhone 15 Pro Max
Augmented Reality Piano Tutoring
PianoVision AR
A remake of an old commercial entirely generated by AI.
The original commercial required 30 people and took about a month to produce, whereas the AI-driven remake was accomplished by one individual in less than a day.
Created with Pika Beta 2.0 (Coming Soon!)
AI enhanced motion capture is wild
The future will undoubtedly be simulated
Cool example of AI compositing – Making an AI cheeseburger for product imagery
Shaping AI image gen through fast, gestural, and painterly 3D turns out to be pretty powerful.. And Dreams is perfect for this! The sculpt in the upper left corner is a 3D sculpt I made in 2 minutes of 3D sketching on my PS5. The video feed from my Playstation was passed through a “Real-Time-Latent-Consistency-Model” that interprets the video while I’m sculpting so I see this happening as I’m making the 3D model.
LumaLabs is Incredible for Augmented Reality – Worth Knowing Them
I’m testing @LumaLabsAI’s fresh off the press text-to-3D model, Genie. Pairing this super power with a drag n’ drop tool like @playbookXR can help
creators/designers/artists/storytellers instantly realize their visions
Gaussian Splatting Is Weird but Compelling
Lighting splats – first steps in bringing light to gsplats
Announcing SuperSplat 📢
A new open-source tool for working with #GaussianSplatting scenes
Mapping Graphics Over a Wireframe – Incredible
This reminds me of the audio cloning examples from last week. It’s a lot easier to take a clone and map it against an existing lattice (aka original vocal) than create the vocal from scratch. The future of creation is hybrid – have an underlying 3D representation you can exert fine grain control over; then use GenAI to take it all the way.
3D Motion Caption – Again by Luma (worth knowing)
3D capture is moving so fast – I scanned & animated this completely on an iPhone. Last summer you’d need to wrangle COLMAP, Instant NGP, and FFmpeg to make NeRFs. Now you can do it all inside Luma AI’s mobile app. Capture anything and reframe infinitely in post!
Mapping a House Before Selling It as A Keepsake
i’m convinced the killer use case for 3d reconstruction tech is memory capture
my parents retired earlier this year and i have immortalized their home forever more
ChatGPT
ChatGPT Does Better If You Add Emotional Appeals to Prompts!?
This was a study I was waiting for: does appealing to the (non-existent) “emotions“ of LLMs make them perform better? The answer is YES. Adding “this is important for my career” or “You better be sure” to a prompt gives better answers, both objectively & subjectively!
ChatGPT Plus members can now upload and analyze files
ChatGPT Plus members can also use modes like Browse with Bing without manually switching, letting the chatbot decide when to use them.
Many startups just died today. Because OpenAI added PDF chat. You can also chat with data files and other document types. We had a wave of products better suited as features rather than stand-alone companies. Wrappers are being squeezed by OpenAI on one side and incumbents on the other. It’s a rough world out there.
Multimodal GPT using a photo of a meal to estimate calorie count
AI Agents (working on your behalf)
An Explainer By My Long-Time Professional Friend Jeremiah
Soon, everyone will achieve ‘Inbox Zero’, here’s how it’s gonna go down:
-AI will read, summarize, and respond on our behalf.
-Messages will indicate if they were actually written by you.
-This means you’ll get a ‘no’ or ‘yes’ faster based on the recipient’s rules.
-My AI agents will converse with your AI agents.
-AIEO (AI Engine Optimization) will emerge to bypass gatekeeper agents.
-A new social network will surface, verifying that only humans are communicating.
Practical Applications/Science/Business
AI could predict earthquakes a week in advance!
Researchers at UT Austin have developed an AI algorithm that correctly predicted 70% of earthquakes a week before they happened during a seven-month trial in China
https://twitter.com/mustafasuleyman/status/1717589405555196059
Protein Folding
A glimpse of the next generation of AlphaFold
https://deepmind.google/discover/blog/a-glimpse-of-the-next-generation-of-alphafold/
Teachers in India help Microsoft Research design AI tool for creating great classroom content
Scientists excited by AI tool that grades severity of rare cancer
https://www.bbc.com/news/health-67264350
MetNet-3: A state-of-the-art neural weather model available in Google products
https://blog.research.google/2023/11/metnet-3-state-of-art-neural-weather.html
AI can diagnose type 2 diabetes in 10 seconds from your voice
LinkedIn passes 1 billion members and launches new AI chatbot to help you get a job
https://www.cnbc.com/2023/11/01/linkedins-new-ai-chatbot-wants-to-help-you-get-a-job-.html
Enterprise LLMs
Meet Izola: Accelerating Forrester’s GenAI Future
Twitter AI Recommendations
Twitter rolls out AI driven “See similar” posts (which is a bit weird, since algorithms have been “AI” forever…. However, the engine driving it is Qdrant. I like the transparency of the backend.
Speech-to-Text
Real-time transcription powered by enterprise-grade Whisper https://www.gladia.io/blog/real-time-transcription-powered-by-whisper-asr
Introducing Distil-Whisper: 6x faster than Whisper while performing to within 1% WER on out-of-distribution data 🚀 Through careful data selection and filtering, Whisper’s robustness is kept and hallucinations reduced
Robots/Embodiment
Eve is Evolving… the robot’s homepage is a lot slicker now.
https://www.1x.tech/androids/eve
Wearables
The Humane AI Pin apparently runs GPT-4 and flashes a ‘Trust Light’ when it’s recording
https://www.theverge.com/2023/10/24/23930669/humane-ai-pin-trust-light-camera
Apple plans hypertension, sleep apnea detection for next watch, plus AI health coach.
Journalism/News Resources
Generative AI for journalists: Discovering what data can do
Generative AI for journalists: Discovering what data can do
Legal/Security/Ethics
Artists Lose First Round of Copyright Infringement Case Against AI Art Generators
While a federal judge advanced an infringement claim against Stability AI, he dismissed the rest of the lawsuit.
Will AI Be a Commodity Available To All – Or a Tool for the 1%?
“Just as the smart phone has got massively cheaper and is now available to billions of people. The very top don’t get a special smartphone, no matter how rich you are. Everyone has the same access to the best phones” Excellent explanation of the benefits and concerns of AI by
@mustafasuleyman #BBCQT #AISummit
Google Bounty Program
Google expands its bug bounty program to target generative AI attacks
Reddit Gives AI the Heisman
‘Reddit can survive without search’: company reportedly threatens to block Google
https://www.theverge.com/2023/10/20/23925504/reddit-deny-force-log-in-see-posts-ai-companies-deals
This new data poisoning tool lets artists fight back against generative AI
The tool, called Nightshade, messes up training data in ways that could cause serious damage to image-generating AI models.
https://www.technologyreview.com/2023/10/23/1082189/data-poisoning-artists-fight-generative-ai/
AI Godfathers Bengio and Hinton: Major tech companies should devote a third of AI budget to managing AI risk
https://venturebeat.com/ai/ai-godfathers-bengio-and-hinton-major-tech-companies-should-devote-a-third-of-ai-budget-to-managing-ai-risk/
U.S. Tightens China’s Access to Advanced Chips for Artificial Intelligence
Chinese tech giant Alibaba launches upgraded AI model to challenge Microsoft, Amazon
Defense Tech Startup Shield AI Raises $200M At $2.7B Valuation
https://news.crunchbase.com/ai/defense-tech-startup-venture-aircraft-shield/
Exclusive: G7 to agree AI code of conduct for companies
https://www.reuters.com/technology/g7-agree-ai-code-conduct-companies-g7-document-2023-10-29/
Max Tegmark- On Alignment
My new TED talk argues that we should (and *can*) keep increasingly powerful AI under human control – if we do our homework and don’t get carried away by hubris!
Technical/Deep Dives
An interesting breakdown of OpenAI’s product strategy:
Make no mistake, OpenAI is building a new kind of computer, beyond just an LLM for a middleware / frontend. Key parts they’ll need to pull it off:
RedPajama-Data-v2
An open dataset with 30 trillion tokens for training large language models
https://www.together.ai/blog/redpajama-data-v2
New Jimmy Apples leaks about OpenAI’s AGI
Shane Legg (DeepMind Founder) – 2028 AGI, Superhuman Alignment, New Architectures
Meta’s Habitat 3.0 simulates real-world environments for intelligent AI robot training
Researchers develop ‘Woodpecker’: A groundbreaking solution to AI’s hallucination problem
Ollama is the easiest way to get up and running with LLMs locally.
Another Local LLM
Introducing http://chatd.ai A new desktop app that turns documents on your computer into conversations, all offline. Powered by @Ollama_ai
Book – “The Coming Wave: Technology, Power, and the Twenty-first Century’s Greatest Dilemma”https://www.amazon.com/Coming-Wave-Technology-Twenty-first-Centurys/dp/0593593952/
Credits/Sources
At least half of these links come from just a few incredible sources. Please follow them:
- Robert Scoble – https://x.com/Scobleizer
- Ethan Mollick – https://www.linkedin.com/in/emollick/
- David Armano – https://www.linkedin.com/in/darmano/
- Alan Thompson – https://lifearchitect.ai/
- The Rundown – https://www.therundown.ai/
- Borriss – https://twitter.com/_Borriss_
- Bilawal Sidhu – https://twitter.com/bilawalsidhu/
- TLDR – https://tldr.tech/ai
Leave a Reply