• Chaos Theory
  • Posts
  • 🥟 Chao-Down #213 Microsoft offers politicians protection from deepfakes, Brain scan AI shows promise for finding Alzheimer's genes, A website puts out bounties for deepfakes of real people

🥟 Chao-Down #213 Microsoft offers politicians protection from deepfakes, Brain scan AI shows promise for finding Alzheimer's genes, A website puts out bounties for deepfakes of real people

Plus, a look at OpenAI's six-member board who will decide "when we've attained AGI"

Last week, OpenAI’s Dev Day took the AI community by storm. Among their various announcements was a new Assistants API that makes it easy for developers to create AI agent experiences.

The Semantic Kernel team (the product I work on) showed off what an initial integration between these two frameworks could look like. Check it out!

-Alex, your resident Chaos Coordinator.

What happened in AI? 📰

Microsoft offers politicians protection against deepfakes (The Verge)

AI that reads brain scans shows promise for finding Alzheimer’s genes (Nature)

ESA - AI maps icebergs 10,000 times faster than humans (European Space Agency)

AI Bot Carried Out Insider Trading and Lied About Its Actions, Study Shows (Business Insider)

OpenAI’s six-member board will decide ‘when we’ve attained AGI’ (VentureBeat)

Giant AI Platform Introduces ‘Bounties’ for Deepfakes of Real People (404media.co)

Always be Learnin’ 📕 📖

Retool State of AI Report 2023 (Retool)

How to use ChatGPT in your design work - by Felix Lee (substack.com)

Adversarial Attacks on LLMs | Lil'Log (lilianweng.github.io)

Projects to Keep an Eye On 🛠

Announcing Together Inference Engine – the fastest inference available (Together.ai)

Deci-AI/super-gradients: Easily train or fine-tune SOTA computer vision models with one open source training library. The home of Yolo-NAS. (Github)

github.com/deepseek-ai/DeepSeek-Coder - Let the Code Write Itself (Github)

The Latest in AI Research 💡

Rethinking Benchmark and Contamination for Language Models with Rephrased Samples (arxiv)

ziqipang/LM4VisualEncoding: Official Implementation of "Frozen Transformers in Language Models are Effective Visual Encoder Layers" (Github)

PJLab-ADG/GPT4V-AD-Exploration: On the Road with GPT-4V(ision): Early Explorations of Visual-Language Model on Autonomous Driving (Github)

Weather Understanding

The World Outside of AI 🌎

Coping With Anger as a Family Caregiver (The New York Times)

Three big reasons Americans haven't rapidly adopted EVs (BBC Worklife)

Can Humans Control the Weather? Japanese Scientists Think They Can (Bloomberg)

Tumblr is downscaling after failing to ‘turn around’ the site (The Verge)

Chinese Laser Sensor Maker Targeted as Tech Trade War Spreads (WSJ)

One Last Bite 😋