• WKND AI
  • Posts
  • OpenAI Just HIRED BACK Their CEO

OpenAI Just HIRED BACK Their CEO

And How To Make Your Own GPTs

Hello WKND AI Warriors!

OpenAI took everyone on a rollercoaster ride.

It was a big week for AI news even without all the OpenAI drama.

Here are some of the notable headlines:

OpenAI releases ChatGPT Voice features to free users, which was previously only available to paid ChatGPT Plus users.

Anthropic released Claude 2.1. A huge move given OpenAI’s debacle. You may remember Claude from one of our previous newsletters.

Inflection AI released their own large language model (LLM) called Inflection-2, which directly competes against ChatGPT, Claude, Google Bard, and Meta AI.

Elon Musk announced Grok will officially be launched end of this month to paid subscribers.

A federal judge dismissed most of Sarah Silverman’s lawsuit against Meta, which may have huge implications for AI-related copyright infringement lawsuits.

And, Stability AI released ‘Stable Video Diffusion’, a huge breakthrough in AI generative art. Read more about it in our ‘AI Tool Of The Week’.

A busy week to say the least.

So, without further ado…

Today’s newsletter includes:

  • 📰 AI Featured News

  • 🤿 AI Deep Dive Topic

  • 🛠️ AI Tool Of The Week

  • 🎨 AI Image Of The Week

AI Featured News

If you recall last week, OpenAI abruptly fired their CEO, Sam Altman, only to later reinstate him.

Here are the highlights.

Thursday (Nov 16): Mira Murati is informed she will be promoted from CTO to interim CEO.

Friday (Nov 17): Ilya Sutskever informs Greg Brockman that he is demoted from the board, and Sam Altman that he is fired. Brockman and three senior researchers resign in solidarity with Altman.

Saturday (Nov 18): Investors call for Altman’s reinstatement, but discussions with the board stall.

Sunday (Nov 19): Altman returns to OpenAI’s offices to negotiate his return with the board. The board stalls and appoints Emmett Shear of Twitch as interim CEO.

Monday (Nov 20): Altman and Brockman announce that they are joining Microsoft, OpenAI’s largest investor. Ilya walks back his participation in ousting Altman. 650 employees write an open letter threatening to resign unless the board resigns and reinstates Altman as CEO.

Here’s the kicker, Ilya signed the open letter. 🙄

Wednesday (Nov 22): The board reached an agreement “in principle” for Altman to return as CEO, and announces new board members.

Even though Sam is back, there are many questions left outstanding.

Which board member ousted Altman?

Why was he fired to begin with?

People thought Ilya was the one who ousted Altman.

Then they thought it was Adam D’Angelo because he recently launched a ‘Bot Store’ on his website ‘Poe’.

Why is this a big deal?

The week prior to his firing, Altman announced the highly acclaimed ‘GPT Store’, which would directly compete against Adam’s ‘Bot Store’.

Last but not least, Helen Toner.

Toner and Altman had a disagreement over a research paper she had contributed to. Altman viewed the paper as having painted OpenAI in a negative light.

The paper was more or less about how OpenAI has been rushing to commercialize AI whereas other competitors such as Anthropic (founded by former OpenAI employees) have specifically held back deploying new products to ensure safety and quality.

Just as the dust was settling on the whole debacle, a final plot twist came to light.

OpenAI researchers had sent a letter to the board regarding their concern of a recent breakthrough the team had made called Project Q*.

The nuts and bolts of it is that Q* can do math at a grade school level.

Which doesn’t seem like a big deal, but suggests that the AI can reason on its own.

And if it can reason on its own, it can become unpredictable over time.

Could Q* be the real reason behind Altman’s ousting?

Could his ousting have all been smoke and mirrors?

A drama to play out before our eyes to distract us from the fact that a major breakthrough could have occurred…Q*…a giant leap towards achieving true AGI.

Time will tell, but the rollercoaster ride appears to be over…for now.

AI Deep Dive

ChatGPT took the world by storm by answering questions on any topic.

But what if you could customize it further for specific uses?

That's exactly what GPTs allow you to do.

GPTs are personalized versions of ChatGPT that you can build in minutes with NO coding.

Simply start a conversation to define its knowledge, capabilities, and tone.

Create a math tutor for your child, a gaming guide for your favorite board game, or a travel agent to help plan your next vacation.

The possibilities are endless when anyone can create a tailored AI assistant.

Also, OpenAI will soon be launching a ‘GPT Store’.

Similar to Apple’s App Store, people will be able to earn money from their creations.

Literally anyone can do it.

So, how can YOU create your own GPT?

First, you must be a ChatGPT Plus member (for $20/month).

From there, its as simple as 1-2-3.

Step 1: Define It

Tell the GPT builder you want to create a funny GPT about cats. Specify it will tell jokes, create memes/images, and have an upbeat, funny tone with some sass.

Step 2: Tailor It

Pick a name like ‘CatGPT’. Generate a cute robot cat logo with DALL-E 3. Define allowed topics and content, like avoiding adult content. Add files for more context, and select capabilities like web browsing and image generation with DALL-E 3.

Step 3: Share It

Publish CatGPT publicly so others can access your funny feline assistant. Consider submitting it to the GPT Store under a "pets" or "humor" category for others to find.

AI Tool Of The Week

Whether you're an AI professional or AI enthusiast, these tools promise to increase your productivity 10x.

🛠️ Send us your favorite AI tool to be featured next week.

Stable Video Diffusion: Animation made easy!

  • AI Video Generation: Create videos from images. Two models are available, generating 14 or 25 frame clips.

  • Customization: Play around with frame rates from 3 to 30 fps. The model is adaptable too, allowing custom training for applications like multi-view synthesis.

  • Early Research Preview: Currently intended for research only while Stability AI refines quality and safety. You can join a waitlist for upcoming new experiences, like text-to-video creation.

AI Image Of The Week

Whether crafted by our team on DALL-E 3 (or Midjourney) or submitted by readers like you, these are sure to inspire.

🎨 Send us your favorite AI image to be featured next week.

Credit: alihustle on Midjourney

seeing clear through glasses looking at city, around the glasses everything is blurry --ar 16:9

Hello there! How'd you like this newsletter?

I'm all ears, your feedback helps me make cooler emails for you!

Login or Subscribe to participate in polls.

To support this newsletter, tell a friend to subscribe here!