• Chaos Theory
  • Posts
  • 🥟 Chao-Down #272 OpenAI releases Sora, their new text-to-video model, Google Gemini introduces Gemini 1.5 Pro with a one-million token context window, Apple plans a rival to Github Copilot

🥟 Chao-Down #272 OpenAI releases Sora, their new text-to-video model, Google Gemini introduces Gemini 1.5 Pro with a one-million token context window, Apple plans a rival to Github Copilot

Plus, Amazon's AGI team claims their AI is showing "emergent abilities.

Have we just witnessed the next evolution of video generation?

OpenAI just released their latest model Sora and showed off some impressive demos that reimagine what’s possible with AI. In contrast to the previous state of the art, Sora is able to generate videos that are up to one-minute long and do so with a high level of fidelity and temporal coherence.

According to OpenAI:

Sora is able to generate complex scenes with multiple characters, specific types of motion, and accurate details of the subject and background. The model understands not only what the user has asked for in the prompt, but also how those things exist in the physical world.

Check out some additional videos that OpenAI generated here.

-Alex, your resident Chaos Coordinator.

What happened in AI? 📰

Amazon AGI Team Say Their AI Is Showing "Emergent Abilities" (Futurism)

Apple’s AI Plans: GitHub Copilot Rival for Developers, Tool for Testing Apps - (Bloomberg)

Google Introduces Gemini 1.5 Pro with a Massive 1 Million Context Window (Beebom)

Who makes money when AI reads the internet for us? (engadget.com)

People are using ChatGPT to write performance reviews (Axios)

Dude, where’s my self-driving car? (The Verge)

Always be Learnin’ 📕 📖

What Is a Design System? | Design Systems 101 | (Figma Blog)

The Impact of Founder Education on Startup Success (datadrivenvc.io)

Synthetic Data for Finetuning: Distillation and Self-Improvement (eugeneyan.com)

Projects to Keep an Eye On 🛠

reorproject/reor: AI note-taking app that runs models locally. (Github)

BCG-X-Official/agentkit: Starter-kit to build constrained agents with Nextjs, FastAPI and Langchain (Github)

adamcohenhillel/ADeus: An open source AI wearable device that captures what you say and hear in the real world and then transcribes and stores it on your own server.(Github)

The Latest in AI Research 💡

Boximator: Generating Rich and Controllable Motions for Video Synthesis (arxiv)

Unified Hallucination Detection for Multimodal Large Language Models (arxiv)

SynthCLIP: Are We Ready for a Fully Synthetic CLIP Training? (arxiv)

The World Outside of AI 🌎

Nandia is repurposing obsolete airplanes into EV parts (qz.com)

Building durable basketball players from the ground up (way up) (Ars Technica)

Why Americans Suddenly Stopped Hanging Out (The Atlantic)

‘It has really gotten out of hand’: wage theft rampant in US construction | US unions | (The Guardian)

Social networks are getting stingy with their data, leaving third-party developers in the lurch (TechCrunch)

Introducing meat–rice: grain with added muscles beefs up protein (Nature)

One Last Bite 😋

The US Patent Office says that only humans can patent inventions!