🥟 Chao-Down #298 Claude 3 beats GPT4 in latest leaderboards, OpenAI shares public first impressions of Sora, Elon Musk makes Grok AI available to all premium X subscribers

Has Anthropic unseated OpenAI? According to the latest from the Chatbot Arena Leaderboard, Claude 3 Opus has eclipsed GPT-4 in community-sourced rankings.

Interestingly, even Anthropic’s smaller and faster Haiku models are showing higher scores than previous versions of GPT4 and Mistral Large. 

We’re in for an interesting next few months as I expect OpenAI, Google, and all the other model providers won’t sit idly by as they try to reclaim the top spot on the leaderboard.

-Alex, your resident Chaos Coordinator.

What happened in AI? 📰

Claude takes the top spot in AI chatbot ranking — finally knocking GPT-4 down to second place (Tom's Guide)

Elon Musk says all Premium subscribers on X will gain access to AI chatbot Grok this week (TechCrunch)

How state lawmakers, election officials are fighting AI deepfakes (StateScoop)

Google AI search tool surfaces scams, malicious links (The Register)

Mathematicians use AI to identify emerging COVID-19 variants (medicalxpress.com)

Sora: first impressions (OpenAI)

Always be Learnin’ 📕 📖

Unlocking hidden value: How niche markets are bigger opportunities than VCs think (Substack)

Statistical Thinking – What Does a Statistical Method Assume? (fharrell.com)

Claude and ChatGPT for ad-hoc sidequests (simonwillison.net)

Projects to Keep an Eye On 🛠

semanser/codel: ✨ Fully autonomous AI Agent that can perform complicated tasks and projects using terminal, browser, and editor. (Github)

jasonppy/VoiceCraft: Zero-Shot Speech Editing and Text-to-Speech in the Wild (Github)

lichao-sun/Mora: Mora: More like Sora for Generalist Video Generation (Github)

The Latest in AI Research 💡

LLMLingua-2: Data Distillation for Efficient and Faithful Task-Agnostic Prompt Compression (arxiv)

TnT-LLM: Text Mining at Scale with Large Language Models (arxiv)

Agent-FLAN: Designing Data and Methods of Effective Agent Tuning for Large Language Models (arxiv)

The World Outside of AI 🌎

The Couples Embracing the DINK Label (WSJ)

QR codes make it easier to steal plane tickets (qz.com)

Ageism Haunts Some Tech Workers in the Race to Get Hired (WIRED)

Colorado to switch on data-fueled speed limit signs (StateScoop)

Parkinson’s Disease Can Now Be Detected Through the Skin (WSJ)

Pregnancy and Childbirth Reshape the Brain in Profound, Sometimes Lasting Ways (Scientific American)

One Last Bite 😋

Be careful with opening email attachments…