It’s been another insane week in the world of AI. While Tesla CEO Elon Musk was debuting hislong-awaited Cybercabthis week (along with a windowless Robovan that nobody asked for), Google’s AI was helping researcherswin Nobel Prizes, Zoom revealedits latest digital assistant, and Meta sent its Facebook and Instagram chatbotsto the U.K.
Check out these stories and more from this week’s top AI headlines.
Google’s AI helped researchers win two Nobel Prizes
The 2024 Nobel Prizes in chemistry and physicswere awarded to researchers with deep ties to Googleon Wednesday. Former Google researcher Geoffrey Hinton won for physics for his work with foundational machine learning techniques that led to the current AI revolution, while DeepMind co-founder Demis Hassabis won for his efforts to decode protein structures.
However, not all of their peers were satisfied with the awards, with one associate math professor arguing about Hinton’s prize, “What he did was phenomenal, but was it physics? I don’t think so. Even if there’s inspiration from physics, they’re not developing a new theory in physics or solving a longstanding problem in physics.”
Adobe is giving creators a way to prove their art isn’t AI slop
It’s getting to where you can’t even believe everything you read and see on the internet these days, what with57% of it likely being AI-generated. To help protect artists (aka its customer base)Adobe is launching a beta of its Content Authenticity web appin the first quarter of next year that will enable content creators to certify that their art is human-made.
This is no simple metadata, mind you — Adobe’s certification system uses a combination of digital fingerprinting, watermarking, and cryptographic metadata to prove the provenance of images, video, and audio files, and can’t be easily stripped or circumvented.
Meta’s AI Chatbot goes to London
After launching in the U.S. and Australia, Meta announced this week thatit is bringing its AI chatbot to Facebook and Instagramusers in the United Kingdom and Brazil, as well as onthe company’s smart glasses. The AI assistant can generate text and still images and, as the company recently revealed, will be trained on data generatedby its bespectacled users.
Google releases Imagen 3 (but only paying subscribers get to generate people)
The latest and greatest of Google’s image generation engines,Imagen 3, dropped this week for all users. The new model offers improved photorealism and fewer artifacts (just look at that cat!). I gave it a quick try and it does what it says it does on the box, allowing me to quickly generate an image from a text prompt, then change and refine that output through subsequent iterations.
Zoom debuts its new customizable AI Companion
Zoom has drastically expanded its product line in recent years from a simple video conferencing app to an entire ecosystem of marketing and collaboration tools. This week the companyrevealed its next-generation AI Companion, a digital assistant that will work seamlessly across Zoom’s various apps to help streamline and automate common business tasks, like transcribing meeting notes, setting agendas, and summarizing reports.
Amazon debuts AI ‘Shopping Guides’ for more than 100 product types
You knowthat AI summarythat Google started sticking at the top of its search results page, the one thateverybody seems to despise? Great news,Amazon is doing basically the same thingfor more than 100 types of products it carries.
Now you’ll be able to learn about the key points and important information about that dog food or camping tent you’re considering buying, alongside product listings filtered for your specific search.
