The Weekly D-AI-gest

Navigate the change in AI Landscape with us

Hey there 👋

Welcome to the eighth and super special weekly newsletter from Unwind AI. Why is it super special? We are absolutely thrilled to announce that we have hit over 1100 subscribers, and we couldn't have done it without you. We endeavor to keep prospering and bringing you the latest and most relevant AI developments with your continued support. The past week has seen some exciting developments in the field, and we can't wait to share them with you. From breakthroughs in natural language processing to advancements in machine learning, there's a lot to catch up on.

Let’s not take more time and start exploring!

This issue covers:

  1. Latest Developments 🌍

  2. News from the Industry 🧑‍🏫

  3. Tools of the Trade ⚒️

  4. Hot Takes 🔥

  5. AI Meme of the Week 🤡

Latest Developments 🌍

Image Editing Simplified with AI 🎨

InstructPix2Pix is a new method for editing images based on human instructions. The method uses a combination of GPT-3 and Stable Diffusion to generate a large dataset of image editing examples. The model can generalize to real images and user-written instructions at inference time, and can perform edits in just a matter of seconds without requiring per-example fine-tuning or inversion. The results are compelling for a diverse collection of input images and written instructions.

Teaser

The Future of Audio Narrations 🎙️

ElevenLabs, an AI voice technology start-up, has launched a beta platform for voice synthesis, cloning, and designing synthetic voices, aimed at providing users with new creative outlets for long-form content.

The company's goal is to become the first AI narrator to provide the quality necessary for voicing news and audiobooks at scale, and also make on-demand multilingual audio support a reality across streaming, audiobooks, movies and real-time conversation.

The above audio is generated using ElevenLabs in the tone “conversational/Rachel (American, mellow).

LOLNeRF: Learn from One Look 👁️

Google has presented LOLNeRF, a framework that allows the user to learn a high-quality representation of 3D objects with just a single 2-D image. Combining Generative Latent Optimization (GLO) and Neural Radiance Fields (NeRF), LOLNeRF provides state-of-the-art results for novel view synthesis and competitive results for depth estimation. This eliminates the need to capture multiple angles, making it an invaluable tool for object recognition.

News from the Industry 🧑‍🏫

OpenAI + Microsoft = LLMs to the moon 🚀

OpenAI and Microsoft have announced an extension to their partnership, with Microsoft investing multi-year, multi-billion dollars. The partnership will allow OpenAI to continue its independent research and develop AI that is increasingly safe, useful, and powerful, while Azure will remain the exclusive cloud provider for all OpenAI workloads across research, API, and products.

The partnership will also involve building supercomputing systems powered by Azure, working together to deploy OpenAI's technology and building OpenAI's technology into apps, and collaborating to review and synthesize shared lessons for the industry.

Microsoft Extends Partnership with ChatGPT Creator OpenAI

ChatGPT is ‘not particularly innovative’ and ‘nothing revolutionary’ 🌶️

Where people still can’t over the wonders of ChatGPT, Yann LeCun, Meta's chief AI scientist, said in an interview that ChatGPT is not particularly innovative and that the idea of OpenAI being alone in its type of work is inaccurate. LeCun stated that similar data-driven AI systems have been built in the past and that OpenAI is not particularly an advance compared to the other labs.

Why hasn't the public seen programs like ChatGPT from Meta or from Google? "The answer is, Google and Meta both have a lot to lose by putting out systems that make stuff up,"

Yann LeCun being the typical him! 😌

yann-lecun-at-collectivei-jan-2023.png

(Source: zdnet.com)

Prompt Engineer’s salary = 3x Software Engineers’ salary 😲

Wrote an article a year ago about the future of prompt engineering and was met with skepticism from many people who thought it to be just hype.

Well, now, prompt engineers are being offered significantly more than software engineers, with companies like Anthropic AI offering salaries of $250-$335k, along with equity. Tap on the link to see the job listing.

Tools of the Trade ⚒️

When Search meets AI 🔍

Andi is a next-generation search engine that uses generative AI and a conversational interface to provide answers to complex questions and provide information from the best sources, rather than just links. It has a conversational interface, like messaging a smart friend, and has features like:

  • A "synthesis engine" for question answering

  • Tools to fight spam and bad content

  • A distraction-free reader view of articles

  • Visual results

It is private, ad-free, anonymous, and free to use.

Go ahead and try it out!

Where AI meets Nocode 🔥

Metatext lets you build and deploy AI models for text classification tasks like text analysis, ticket routing, review analysis, spam detection, topic detection, etc. without any machine learning expertise. It provides an easy-to-use UI, APIs, and pre-trained models, allowing users to develop custom AI models and automate complex workflows on-the-fly.

Metatext Build AI without code

Delicious Dishes by ChefGPT 🧑‍🍳

You don’t have to struggle with discovering new recipes for yourself because ChefGPT has got you covered! ChefGPT uses AI to suggest recipes based on the ingredients and tools available to you, as well as other factors such as time constraints and cooking skill level. You can also customize their recipe suggestions by selecting different "Chef Modes" like Gourmet Mode and All-In Mode.

[crop output image]

Spam Fighting Sidekick 🥊

Outwit spammers with AI-generated replies. Begone Spammer will help you outsmart spammers by generating personalized, snappy replies. You only have to paste the unsolicited message and choose the “Reply Flavour” like Begone or Mad Max and relax!

Going Beyond 2D 🧱

Genmo is a platform for creating and sharing interactive, immersive generative art with a single click. It allows users to go beyond 2D images and create videos, animations, vector design assets, and more. The platform also allows advanced video options like dynamism, which controls how much noise is added to video sampling between frames and chaining multiple prompts to tell a story.

Hot takes 🔥

I thought AGI was so yesterday 😂

Wharton MBA or ChatGPT pro?

And the list goes on…

AI Meme of the Week 🤡

Image

That’s all for this week!

Will see you next Saturday with more such content. Don’t forget to subscribe and give your feedback below!

BONUS 🎉

Share this newsletter with three other friends and stand a chance to win my book GPT-3: The Ultimate Guide to build NLP Products with OpenAI API. Winners will be selected on a monthly basis.

🎁 Every paid subscriber will also receive $39 USD worth of learning resources on trending topics like Python, Data Science, Machine Learning, and NLP! 

Reply

or to participate.