AI reading list #4 on GPT Store, Data Ownership, Reinforcement Learning and Model Transparency

This week, I couldn't miss the excitement surrounding the GPT Store's next launch. This is going to be a hit. Let's see how GPTs perform and how the final monetization model will evolve // Data also fascinates me in the context of General AI, and I'm bringing an article about how we are entering the Data Economy // You have probably heard about Reinforcement Learning, but the opportunities presented by General AI are transforming the entire learning sector // Transparency and Ethics have also been main topics in this week's AI and General AI-related news.

• The GPT store will launch next week (via OpenAI)
• Data ownership is leading the next tech megacycle (via Techcrunch)
• Revolutionizing Education: 5 Eye-Opening Stats on How Reinforcement Learning Transforms Student Success (written by Chitranshu Harbola)
• Introducing The Foundation Model Transparency Index (via Standford-HAI)

The GPT store will launch next week (via OpenAI)

This was a hot topic for GPT developers, and the time has come for all of us to launch our GPTs in the store. Be aware that this big move is not just for software developers; anyone can create and train a GPT and apparently monetize a GPT.

If you want to enter this world of Gen AI, this is a good way to start playing with models and understanding how they work, why feeding data is important, and why we must be careful with the responses we get. You don't need to know about technical stuff (promised).

If you still need to fine-tune your prompts, here you have the full prompt guide. And if you want to explore some prompt examples, you can check them out here.

Read article: https://l.getvond.com/7caee4

Data ownership is leading the next tech megacycle (via Techcrunch)

The Data Economy has arrived. The reality that AI and Gen AI are data-driven processes is undeniable. How they will evolve remains uncertain, but this article suggests some intriguing paths ahead. Our perspectives on data — how we protect, store, or retrieve it — are all being reconsidered.

Recent events offer illustrative examples of this data-centered approach: The New York Times initiated a lawsuit against OpenAI, alleging that its algorithms were trained using the newspaper’s data. In response to such challenges, OpenAI is proactively taking steps to reduce regulatory risks in the EU, particularly concerning data privacy.

Read article: https://l.getvond.com/2d0528

Revolutionizing Education: 5 Eye-Opening Stats on How Reinforcement Learning Transforms Student Success (Written by Chitranshu Harbola)

I have recently been reading about reinforcement learning and AI and the opportunities it presents for completely transforming educational processes. While the concept isn't new, it certainly enters a new dimension with the possibilities that General AI offers to tailor each learning concept to suit each student’s unique experience. This could lead to a kind of super-adaptive training from which all of us could benefit, not only in schools but also in our companies.

If you are interested in the learning applications of AI and General AI, Fengchun Miao, UNESCO’s Chief of the Unit for Technology and AI in Education, is one of my favourites. If you know more profiles to follow, please share.

Read article: https://l.getvond.com/4065a5

Introducing The Foundation Model Transparency Index (via Stanford-HAI)

A few weeks ago, I introduced the concept of a Foundation Model. Now, let's delve a bit deeper. The Center for Research in Foundation Models has launched an interesting approach to transparency. In this article, you can see how they have developed the scoring model and how some of the most well-known models, such as Llama 2, GPT-4, Claude 2, and others, are performing. The results may not be promising, but they are certainly worth considering.

Read article: https://l.getvond.com/85ab79