Getting Started

The quickest way to get up to date on what’s going on is probably to read OpenAI’s 2023-03-14 GPT-4 Announcement

News

It feels like there’s a year’s worth of announcements and launches per week, and it’s hard to keep up, but here are some ways to try…

News Site

Weekly Newsletters

Leonard’s Firehose Feeds

Products

Top

These are the things that are free and easy to try…

ChatGPT

Bing Chat

Bing Image Creator

Midjourney

Google Bard

Lists

Libraria

Evidence Hunt

ChatPDF

Text to Speech

Eleven Labs

Speechify

NaturalReader

Image Generation

Midjourney

Midjourney V5 is probably the most advanced/best quality AI image generator publicly available atm

InvokeAI

Easiest way to run Stable Diffusion on your local computer

DiffusionBee

On Mac, you can also give this a try:

Design Helpers

Adobe Firefly

https://www.adobe.com/sensei/generative-ai/firefly.html

Imgcreator

https://imgcreator.ai/

Video

RunwayML

Provides a suite of tools for video generation, modification, replacement, etc.

NeRF

Luma

https://lumalabs.ai/

Nvidia Instant NeRF

https://github.com/NVlabs/instant-ngp https://developer.nvidia.com/blog/getting-started-with-nvidia-instant-nerfs/ https://www.nvidia.com/en-us/research/ai-art-gallery/instant-nerf/

Coding

Github

Notes and Productivity

Notion AI

Mem.ai

https://get.mem.ai/

Microsoft Loop

Microsoft 365 Copilot

Transcription

AssemblyAI

https://www.assemblyai.com/playground/source

Whisper

https://openai.com/research/whisper https://github.com/openai/whisper https://github.com/ggerganov/whisper.cpp

ORG

Music

Research

AGI

2023-03-22 Sparks of Artificial General Intelligence: Early experiments with GPT-4

2023-02-04 Theory of Mind May Have Spontaneously Emerged in Large Language Models

Projects

Self Hosted

Lists

Web Scraping

ScrapeGhost

https://jamesturk.github.io/scrapeghost/ scrapeghost is an experimental library for scraping websites using OpenAI’s GPT.

Custom Search

Markprompt

https://markprompt.com/ https://github.com/motifland/markprompt Markprompt is a platform for building GPT-powered prompts. It scans Markdown, Markdoc and MDX files in your GitHub repo and creates embeddings that you can use to create a prompt, for instance using the companion Markprompt React component. Markprompt also offers analytics, so you can gain insights on how visitors interact with your docs.

Text Annotation

Potato

https://github.com/davidjurgens/potato

Video

Text2Video-Zero

ModelScope Text to Video Synthesis

https://huggingface.co/damo-vilab/modelscope-damo-text-to-video-synthesis https://modelscope.cn/models/damo/text-to-video-synthesis/summary https://twitter.com/_akhaliq/status/1639434628166459395 https://www.reddit.com/r/StableDiffusion/comments/120j6h8/barry_chuckle_absolutely_shredding_in_front_of_an/

Voice

VALL-E

Misc

AI Safety

https://cdn.openai.com/papers/gpt-4-system-card.pdf https://langchain.readthedocs.io/en/latest/modules/chains/examples/constitutional_chain.html https://www.alignmentforum.org/

https://twitter.com/goodside/status/1639136386404982784

Pre-2008: We’ll put the AI in a box and never let it out. Duh.
2008-2020: Unworkable! Yudkowsky broke out! AGI can convince any jail-keeper! 2021-2022: yo look i let it out lol
2023: Our Unboxing API extends shoggoth tentacles directly into your application [waitlist link]

https://twitter.com/lxrjl/status/1639397697084874752

"Why would you think AI might end up displaying [deception/power-seeking/other scary thing]?"
"People will design them to"
"But those theorems might not apply to the real wo.... WAIT WHAT?"
"People will design them to"

https://twitter.com/davidad/status/1639215289677017099

AGI

Time to AGI

https://www.metaculus.com/questions/3479/date-weakly-general-ai-is-publicly-known/ https://twitter.com/wintonARK/status/1639291591549681666

https://en.wikipedia.org/wiki/Existential_risk_from_artificial_general_intelligence https://www.lesswrong.com/posts/gEchYntjSXk9KXorK/uncontrollable-ai-as-an-existential-risk https://en.wikipedia.org/wiki/Superintelligence

https://www.anthropic.com/index/core-views-on-ai-safety

https://twitter.com/AnthropicAI/status/1603791161419698181

Sparks of AGI: https://www.youtube.com/watch?v=Mqg3aTGNxZ0

https://twitter.com/DV2559106965076/status/1638769434763608064

https://arxiv.org/abs/2303.12712

https://nlpurr.notion.site/Sparks-of-Artificial-General-Intelligence-Early-experiments-with-GPT-4-d10b39b0e65e40f2b7e23365dd5af59e

https://www.lesswrong.com/posts/azRwPDbZfpadoL7WW/an-appeal-to-ai-superintelligence-reasons-to-preserve

AI Tool Use

https://openai.com/blog/chatgpt-plugins https://www.reddit.com/r/dataengineering/comments/10ijsqy/i_built_an_llmpowered_tool_that_can_understand/ https://til.simonwillison.net/llms/python-react-pattern

Prompt Engineering

https://lifearchitect.ai/bing-chat/#prompt https://www.jailbreakchat.com/ https://www.reddit.com/r/aipromptprogramming/

Training Costs

https://simonwillison.net/2023/Mar/17/beat-chatgpt-in-a-browser/ https://simonwillison.net/2023/Mar/13/alpaca/

$30 https://twitter.com/alighodsi/status/1639251347777388544

https://twitter.com/neuralmagic/status/1638889340020625409

Examples

Fiction

Learning

  • LLM Introduction: Learn Language Models
    • “Bootstrap knowledge of LLMs ASAP. With a bias/focus to GPT. Avoid being a link dump. Try to provide only valuable well tuned information.”
  • AI Notes: Top AI Reads
    • “notes for software engineers getting up to speed on new AI developments. Serves as datastore for lspace.swyx.io writing, and product brainstorming, but has cleaned up canonical references under the /Resources folder.”
  • Simulators
    • “TL;DR: Self-supervised learning may create AGI or its foundation. What would that look like?”