Big Reveals from Top AI Contenders: A Day Full of Exciting Announcements

Yesterday was a huge day for AI large language models (LLMs). All the major players had big announcements...keep on reading to get caught up!

We'll give you a quick summary of every major announcement and dive into each one more thoroughly in future newsletters. This way, we can focus on the specifics of each item rather than skimming through them based on how they stack up against the others.

Here's what we got for you today:

• 🤖 OpenAI GPT-4

• 📑 AI Features for Google Workspaces

• 🎱 Anthropic Claude

• 💬 New Bing Is Running GPT-4

• 🚀 Cool A.I. Tools, Demos, and News

• 🖼️  A.I. Generated Image of the Day

• ⚡️ A.I. Meme of the Day

GPT-4 Released!

GPT-4 released today and is available to ChatGPT Plus subscribers only. User are capped at 100 messages every 4 hours for now. OpenAI will adjust the cap for GPT-4 usage based on demand. When logging into ChatGPT Plus at ai.com you should see invite to try out GPT-4. If you don’t see it give it some time, OpenAI is slowing rolling out this option to all Plus subscribers. Click here for details about announcement and click here for the research that went to GPT-4.

Below is a video from GPT-4 Developer Livestream where Greg Brockman, President and Co-Founder of OpenAI demonstrates some of the features.

AI Features for Google Workspaces

Amidst the buzz about GPT-4, Google's announcement about integrating AI across their entire Workspace – Gmail, Docs, Sheets, Slides, Images, and more – seemed to fly under the radar. In many ways, this announcement is just as significant as GPT-4, but Google's presentation is light on details and heavy on marketing language. Take a look for yourself here and here, and let me know what you think. Short recorded demo marketing video below.

Anthropic Claude Released!

Another big announcement - Anthropic, the AI company started by former OpenAI team members, has finally released its AI chatbot named Claude. While Claude performs many tasks similar to OpenAI's ChatGPT, early users claim it's less likely to generate harmful content and is more user-friendly when it comes to conversation. More info about Claude here. If you can’t get access to Claude via the request form, you can always use it via Poe.com.

New Bing Is Running GPT-4

With the release of GPT-4 all Bing / Bing Chat is powered by latest GPT-4 model. It was also revealed that early users of Bing Chat were using an early model of GPT-4! More info here and the waitlist link. If you are already have early user access you don’t need to sign up for waitlist.

🚀 Cool A.I. Tools, Demos, and News

Here are some of the most exciting AI-powered tools and news from the last few days.

🖼️ 100% A.I. Generated Image of the Day

Midjourney Prompt - editorial photo from Dwell, Midcentury modern house, on a cliff overlooking Los Angeles, evening sun, brilliant architecture, beautiful, exclusive, expensive, minimal lines, breathtaking, 8K, architecture photography --ar 3:2

🤭 A.I. Meme of the Day

OpenAI and GPT-4 are stealing the spotlight this week, but we believe the other announcements are very important too, and they'll likely gain more attention in the weeks ahead.

🔬 A.I. / Machine Learning Research

🌴 PaLM-E - What happens when we train the largest vision-language model and add in robot experiences? The result is PaLM-E , a 562-billion parameter, general-purpose, embodied visual-language generalist - across robotics, vision, and language.

🧐 Visual ChatGPT - One of the most significant breakthroughs is ChatGPT, which is built upon InstructGPT. Although powerful, ChatGPT is limited in its ability to process visual information since it is trained with a single language modality, while Visual Foundation Models (VFMs) have shown tremendous potential in computer vision, with their ability to understand and generate complex images.

🧮 MathPrompter -Large Language Models (LLMs) have limited performance when solving arithmetic reasoning tasks and often provide incorrect answers. Unlike natural language understanding, math problems typically have a single correct answer, making the task of generating accurate solutions more challenging for LLMs.

👣 GigaGAN - GigaGAN offers three major advantages. First, it is orders of magnitude faster at inference time, taking only 0.13 seconds to synthesize a 512px image. Second, it can synthesize high-resolution images, for example, 16-megapixel pixels in 3.66 seconds.

♻️ Refer Daily Zaps

Can you do me a solid and refer a buddy, colleague, or family member to Daily Zaps? As a thank you, I'll send you an Airtable database / list of over 170 AI-focused startups complete with links to their sites, founder names, funding amounts, investors, and all that good stuff. Sound good? Let's do this!

Peace out,

Daily Zap Team