Introducing the AI Industry Catalog

Hey Readers, Happy Sunday! 👋

In today's edition, I'll cover the following:

1. My concept of the week: The AI Industry Catalog

   - AI Supertutor from Khan Academy

   - AI Open Source is taking the lead

   - Meet PI - The Emotional Support AI Companion

   - White House Unveils Initiatives to Reduce Risks of AI

   - Samsung Limits AI Tools Over Security

2. My new Interview: A demo of Predictive Maintenance Use Case with Yana, a Senior Data Scientist from IBM

Let's dive in! 🤿

📚 My AI posts this Week: AI Industry Use Cases Catalog

I'm creating the largest catalog of Industry AI Use Cases and I post almost daily new ones. AI is revolutionizing the global economy and as industries adapt to an increasingly digital world, AI continues to create new opportunities and challenges, reshaping the way we live, work, and interact.

Check out the first four use cases that I published this week. I include demos and samples on how to executute them. Let me know if you have any questions or would like to contribute!

• Healthcare: Diagnostics and Medical Imaging

• Manufacturing: Predictive Maintenance

• Energy: Accurate Demand Forecasting

• Real Estate: Property Search

🗞️Top News of the Week

1. Khan Academy AI Super Tutor

In a recent TED talk, Sal Khan introduced Khan Academy's AI bot, Khan Amigo. Unlike other chatbots, Khan Amigo focuses on guiding students through their thought processes rather than providing answers directly. As a "supertutor," it covers various subjects and can act as a guidance counselor or academic coach. It also helps teachers plan lessons, freeing up time for individual student attention. With Khan Academy being a leading online learning resource, this innovation further enhances its educational impact.

2. Document Leak: AI Open-Source Revolution

In a leaked internal document, a Google researcher suggests that open-source models, rather than OpenAI or Google, will lead the AI race. Open-source models have been rapidly improving, with recent developments like Meta's LLaMA offering impressive performance at lower costs. The researcher predicts a surge in faster, more customizable open-source models, which could potentially outpace proprietary models. This development could benefit consumers and the ecosystem by providing more choice, but might pose challenges for businesses relying on proprietary research. Check out parts of the document here. It is a great read.

I will further write about this topic, highlighting the limitations of massive large models in terms of flexibility and the undeniable influence of open-source innovation across various fields. As we investigate this, I'll explain how the open-source movement is on the brink of reshaping the AI landscape, fostering a more dynamic and engaging future for the industry.

3. PI - The Emotional Support AI Companion

The AI start-up Inflection AI recently launched an emotional support chatbot named Pi, designed to provide digital companionship and validation. Pi is part of a trend in which AI technologies are being fine-tuned to engage in conversations with users. While some chatbots focus on answering queries or improving productivity, tech companies are increasingly infusing them with personality and conversational skills. Critics warn that AI companionship can create problems if bots give bad advice or enable harmful behavior, and concerns about privacy arise due to the sensitive nature of conversations. Inflection AI aims to build honest and trustworthy AI, and Pi is designed to express uncertainty and recognize its limitations.

4. White House Unveils Initiatives to Reduce Risks of AI

The Biden-Harris Administration is taking new actions to promote responsible AI innovation in the U.S. and protect people's rights and safety. Building on previous initiatives, the administration is emphasizing the importance of responsible AI development that serves the public good and safeguards society, security, and the economy. As part of this effort, Vice President Harris and senior officials will meet with CEOs of Alphabet, Anthropic, Microsoft, and OpenAI.

The Administration has already made progress through the AI Bill of Rights, AI Risk Management Framework, and a National AI Research Resource. New announcements include $140 million for the National Science Foundation to launch seven National AI Research Institutes, a public assessment of existing AI systems by leading developers, and draft policy guidance on AI use by the U.S. government. These actions aim to mitigate AI risks, harness opportunities, and set an example for responsible AI development and use.

5. Samsung Restricts Employee Use of Generative AI Tools Amid Security Concerns

Samsung is temporarily restricting employees' use of generative AI tools like ChatGPT after discovering instances of misuse, including uploading sensitive code. The South Korean tech giant advised employees to exercise caution when using ChatGPT and similar products outside of work, and to avoid entering personal or company-related information. A recent survey found that 65% of Samsung employees expressed concern about security risks when using generative AI services. Other companies like JPMorgan and Amazon have also limited or warned against the use of such tools. Despite these restrictions, Samsung is exploring ways to use generative AI safely to enhance employee productivity and efficiency.

🎙️ Interview of the week: Predictive Maintenance Demo

Yana, a senior data scientist from IBM, was kind enough to share a demo of a Predictive Maintenace solution built with IBM Watson technology. In this demo, you will see all the steps from the prediction of failures for aircrafts and then the creation of optimization scenarios to efficiently plan maintenance.

See you next Sunday!

Armand 😎