Learning from Mistakes in the Tech World [Newsletter #39]

Hello, AI enthusiasts from all around the world! Welcome to the weekly newsletter for the AI and The Future of Work podcast.

The tech world is no stranger to hype. In fact, we're living the AI hype right now. 

There's no shortage of investment in AI-driven companies at the moment, which can make these companies feel invincible. 

However, that very same hype can also blind those looking to create the next life-changing product to the warning signs. 

Today's guest reminds us that the past has seen similar hype cycles, and it leaves valuable lessons. 

Let's dive into this week's highlights! 🚀

🎙️New Podcast Episode with Mona Sabet, Co-founder at ChIPsNetwork.org and Co-author of "Sail to Scale: Steering Startups Clear of Mistakes from Launch to Exit."

Mona Sabet has seen founders make a lot of mistakes.

And all these mistakes have one common trait: they can be fatal for startups. And with these early-stage companies, it’s generally not a question of if, but when it will kill them.

However, just as Mona has seen many founders make these mistakes, she has also seen countless success stories.

Many of them follow a straightforward recipe: having strong leadership principles and equally strong product management.

Still, the missteps keep happening, across industries. So Mona partnered with two other experienced founders and co-wrote "Sail to Scale: Steering Startups Clear of Mistakes from Launch to Exit," which launched in August 2024.

Mona is also the current Senior Vice President of Strategy, Projects, and Operations at GCG. Her previous experience includes Chief Corporate Strategy Officer at UserTesting, where she helped close multiple acquisitions while also leading the company's $100M funding round, IPO, and $1.3B acquisition.

Our CEO, Dan Turchin, sat down with Mona to discuss her ample experience in the startup world and her efforts to include women in tech, law, and policy through the ChIPs Network.

In this conversation, Mona and Dan discuss this and much more:

• There's one common misconception founders have about their MVP — and when they realize it, it’s often too late..

• Many leaders fail to admit (even to themselves) that there's a false commitment in their team.

• Mona describes the most common ups and downs of selling startups and how founders can get it right the first time.

• Hype cycles seem unavoidable, but when the current AI hype bubble bursts, the industry will see a major shakeup.

• Everyone must play their part and change how they approach diversity in tech to make the progress the industry demands.

Our latest AI and the Future of Work podcast episode featuring Mona Sabet inspired this issue.

🎧 Listen here to learn more about Mona's critical analysis of the startup world, which industry currently catches her attention, and the one demographic that needs to give itself more opportunities.

📖 AI Fun Fact Article

World simulators sound like something straight out of science fiction, but they are touted as the next big thing in AI. Kyle Wiggers explains what they are and why they matter in TechCrunch. 

The concept of world simulators or 'world models' might be hard to grasp at first, but we can understand them by looking at how our mind works. 

Our brains take abstract representations from our senses—from touch to taste to sound, and turn them into more concrete ideas or models, as Wiggers refers to them. 

With this information, our brains make predictions. These predictions shape how we view the world, and the same can be said about world models. 

This approach could help solve a current dilemma with AI. A generative model can consume years of video to accurately predict how a basketball bounces. 

However, the generative model doesn't actually understand why this happens. It's the same with language models and how they don't really grasp the concept behind words and phrases. 

That's where a world model comes in. Even one with basic knowledge of why the basketball bounces can reason about the consequences of actions.

This possibility has motivated many, and as a result, funding for large world model builders is growing rapidly. AI godmother Fei-Fei Li's company World Lab raised $250 million to build these worlds.

Google's DeepMind hired one of the top minds at OpenAI to help create world simulators. 

Wiggers' article reminded our CEO, Dan Turchin, of his conversation with Peter Voss, who said that world models could very well be the path to "artificial general intelligence" (AGI). 

However, Dan also highlights that we won't get to AI that understands the world with larger generative, pre-trained language models no matter how long we allow them to think.

We'd love to hear your thoughts on this: did you know about world models? Are you excited about their potential? 

Here's the link to the full article if you want to learn more! 

🎶 Worth a Listen! 

👋 That’s a Wrap for This Week!

If you liked this newsletter, share it with your friends!
If this email was forwarded to you, subscribe to our Linkedin’s newsletter here to get the newsletter every week.