- The Rundown AI
- Posts
- Meta AI goes Hollywood
Sign Up | Advertise | Tools | AI University
Welcome, AI enthusiasts.
Meta is making moves in Hollywood, offering celebrities millions to use their voices for its new AI assistant project.
Are we entering a new world with billions of celebrity-powered AI voice agents, or heading for another ScarJo/OpenAI-style controversy? Let’s get into it…
In today’s AI rundown:
Meta AI’s new Hollywood hires
Google absorbs Character AI talent
Large 2: A free-to-use GPT-4 competitor
Tesla unveils new AI vision for robots
6 new AI tools & 4 new AI jobs
More AI & tech news
Read time: 4 minutes
LATEST DEVELOPMENTS
META
Image source: Getty Images
The Rundown: Meta is reportedly offering millions to celebrities like Awkwafina, Judi Dench, and Keegan-Michael Key to use their voices in upcoming AI projects.
The details:
The AI voices would be used across Meta's platforms, including Facebook, Instagram, and Meta Ray-Ban smart glasses.
Meta is reportedly rushing to secure deals before its Meta Connect conference in September.
Contracts are reportedly temporary, with actors having the option to renew.
Meta has previously experimented with celebrity-inspired chatbots, though that program has ended.
Why it matters: In our exclusive interview with Mark Zuckerberg, he predicted that “we're going to live in a world where there are going to be hundreds of millions or billions of different AI agents”. If it holds true, celebrity voice-powered AI could be part of Meta's next big play to drive user engagement and growth on the platform.
TOGETHER WITH ZAPIER
The Rundown: Zapier is hosting its annual virtual user conference on September 26, 2024. Four content tracks, including a special keynote, will offer insights on using AI and automation you don’t want to miss.
Get exclusive access to:
How AI and automation can help you revolutionize your workflows
Future of work insights from Zapier CEO Wade Foster and HubSpot CTO Dharmesh Shah
The latest Zapier features with insider tips from automation experts
Click here to secure your spot to ZapConnect 2024 today.
Image source: Ascannio / Shutterstock
The Rundown: Google has signed a non-exclusive licensing agreement with AI startup Character AI for its large language model technology, while also reabsorbing the startup's co-founders and key talent back into its AI team.
The details:
Character AI co-founders Noam Shazeer and Daniel De Freitas return to Google, their former employer.
Google gains a non-exclusive license to Character AI's language model technology.
About 30 of Character AI's 130 employees, mainly those working on model training and voice AI, will join Google's Gemini AI efforts.
Character AI will switch to open-source models like Meta's Llama 3.1 for its products, moving away from in-house models.
Why it matters: This deal highlights the intensifying race to secure top AI talent, mirroring Microsoft’s recent deal with Inflection and Amazon’s deal with Adept. As AI becomes increasingly critical to tech companies' futures, these talent grabs could reshape the landscape, while raising antitrust concerns.
AI TRAINING
The Rundown: Mistral AI recently released Mistral Large 2, an open-source AI model that rivals GPT-4 in performance across a few key benchmarks, available for free and unlimited use.
Step-by-step:
Visit Mistral AI Chat and create an account for free.
Find the "Model" dropdown menu in the chat interface.
Select "Large 2" from the list of available models.
Start interacting with Mistral Large 2 by typing your prompts or questions.
PRESENTED BY ASSEMBLY AI
The Rundown: AssemblyAI's developer-approved API offers the most AI speech capabilities on the market, including speech-to-text, speech understanding, and full access to Claude 3 models.
AssemblyAI leads the market with:
Industry-leading accuracy at 95% and WER at 4.8%
30% less hallucinations than competitors
Robust capabilities including access to Claude 3 models for in-depth call analysis, summarization, and more
TESLA
Image source: Tesla
The Rundown: Tesla just filed a patent for an AI-powered vision system that could transform how autonomous robots perceive and navigate their environment using only camera inputs.
The details:
The system uses a single neural network to process camera data and output detailed 3D environment information without LiDAR or radar.
It divides space into 3D voxels, predicting occupancy, shape, semantic data, and motion for each in real time.
The tech is designed to run on a robot's onboard computer, enabling immediate decision-making.
This system could be implemented in both Tesla's vehicles and humanoid robots like Optimus.
Why it matters: By relying solely on camera inputs and onboard processing, Tesla's new vision system could enable robots to navigate diverse environments more efficiently and adapt to changes in real time. This would eliminate the need for extensive pre-mapping and accelerate the arrival of affordable, autonomous robots.
NEW TOOLS & JOBS
🎨 Black Forest Labs FLUX.1 - Generate high-quality from prompts
🎵 Neutone Morpho - Transform sounds into auditory experiences in real-time
👨💻 Llama coder - Generate full React apps with Claude Artifacts and Llama 3.1
🔍 Felo Search - Multilingual AI search engine
🤗 mgamate - Supportive 1-on-1 AI copilot for managers
📈 Clarity - Customer-focused sales with AI meeting recording and tracking
💳 Luma AI - Senior Backend Engineer - Payments
👨💻 Limitless AI - Principle Design Engineer
💼 Cohere - Head of Product Marketing - AI & Language Models
📅 OpenAI - Workplace Events Coordinator
QUICK HITS
Adobe Illustrator introduced new AI-powered generative vector features, allowing users to create complex vector art and patterns from text prompts.
UC Berkeley unveiled the 'Berkeley Humanoid', a 2 ft. 9 in. humanoid research platform designed for learning-based control and robust locomotion over various terrains.
Britain canceled $1.7 billion worth of computing infrastructure projects, including plans for an exascale computer, dealing a setback to the country's AI ambitions.
China's Cixin Technology unveiled the Cixin P1, a new 6nm Arm-based 12-core AI chip for PCs, capable of 45 TOPS (trillion operations per second) and compatible with multiple operating systems.
Microsoft modified its AI partnership with UAE-based G42, opting to lease AI products instead of transferring sensitive technology amid concerns over G42's ties to China.
An autonomous robot dentist performed the world's first human dental procedure, preparing a tooth for a crown in about 15 minutes — significantly faster than human dentists.
THAT’S A WRAP
SPONSOR US
Get your product in front of over 600k+ AI enthusiasts
Our newsletter is read by thousands of tech professionals, investors, engineers, managers, and business owners around the world. Get in touch today.
FEEDBACK
How would you rate today's newsletter?Vote below to help us improve the newsletter for you. |
If you have specific feedback or anything interesting you’d like to share, please let us know by replying to this email.
Reply