💡 OpenAI Launches Sora

Plus: Google Reveals Gemini 1.5

Read time: 2 minutes (Click underlined topics or headings for links.)

Hey Edge Family, good morning! OpenAI and Google launched new AI models on the same day!

The Breakdown

💎 3 Top AI Tools

Visme: Generate customizable designs with help from text prompts.

Runway: AI research company shaping the next era of art

Future Tools: AI tools directory with over 1000 tools

🚨 Latest Developments 

Source: OpenAI

Big Takeaway: OpenAI unveiled "Sora," a new AI tool capable of generating 60-second, highly realistic videos from simple text prompts. (See examples)

  • The technology behind "Sora" comes from OpenAI's DALL-E, using AI models trained on video and images to respond to user prompts.

  • The technology is currently being "red-teamed" by a select group of researchers and experts to identify potential misuse and errors. Once it is available for public use, we will let you know.

  • Other companies, including Google with "Lumiere," Meta with "Emu," and the AI video startup Runway, have developed similar text-to-video AI generators, but so far Sora has been noted for its superior length and quality.

Are you going to test out Sora?

(Select one)

Login or Subscribe to participate in polls.

Source: Google

Big Takeaway: Google introduced Gemini 1.5, the next-generation AI model, with improvements in performance and efficiency and a breakthrough in long-context understanding capable of processing up to 1 million tokens.

  • Gemini 1.5’s new model architecture is based on a Mixture-of-Experts (MoE) technique that allows for more efficient training and serving.

  • Gemini 1.5 Pro can process and analyze vast amounts of information, including extensive documents, lengthy videos, and complex codebases.

  • In testing, Gemini 1.5 Pro outperformed the Gemini 1.0 Pro model on 87% of benchmarks and showed comparable quality to Gemini 1.0 Ultra.

🔎 Extra Insights 

Meta AI researchers created V-JEPA, a new model trained using masked video to learn like large language models from text. It constructs an internal picture of the world and recognizes object interactions by guessing missing video segments. This video pretraining could make AI training more efficient, allowing more people to develop huge models.

Apple recently acquired the domain name iWork.ai, sparking speculation that it may highlight AI features in its iWork suite of apps, Pages, Numbers, and Keynote. But Apple often buys domains without using them so for now this is just speculation.

A recent study found that 66% of Americans have seen “altered images” online. Clarity, a startup, uses AI models to scan media and compare it to a library of deepfakes, watermarking to verify authenticity. Walden Catalyst and Bessemer led a $16 million seed investment to extend their platform.

💻 Real Life Use Cases

Chat GPT Prompt: Latest Trends Twitter Thread

Use this ChatGPT prompt to create a mock Twitter thread for the latest trends in any industry.

Copy and paste the prompt below:
Start a thread discussing the latest trends in (Insert industry or topic, e.g., 'digital marketing'). Each tweet should highlight a different trend and its implications for (Insert specific group or field, e.g., 'small businesses'). 

Want to access 500+ more research-backed prompts like this? Join The Edge Pro Today! *Special early bird offer: 35% off!

Here’s a link to the results:

💼 Who’s Hiring in AI

Job Title

Company

Salary Range

Location

AI/ML Engineer

Keysight Technologies

N/A

On-site

AI Tutor

xAI

$35-$65/hr

Remote

Head of Sales

Robust Intelligence

$200k-$250k

On-site

🎲 AI Game

Which Picture is AI Generated?

(Select one)

Login or Subscribe to participate in polls.

How was today's newsletter

Login or Subscribe to participate in polls.

That’s all for today!

Thinking of starting your own newsletter? Join Beehiiv, the #1 newsletter platform for built-in growth and monetization! Use my link for a special gift.

If you have something interesting or just want to reach out, don’t hesitate to DM me on Twitter @Brandoncarterss

Subscribe to keep reading

This content is free, but you must be subscribed to The Edge to continue reading.

Already a subscriber?Sign In.Not now

Join the conversation

or to participate.