Blog Series: Exploring the Power of AI

This is part 5 in an ongoing series to acquaint people with AI.

Part 1: The AI Revolution: Why You Should Care and Why You Should Trust Us

Part 2: Demystifying AI: What It Is and Why It Matters

Part 3: Intelligent Innovations: AI in daily life

Part 4: The Future of Work: AI’s Role in Professional Settings

Introduction

It’s hard to identify resources for artificial intelligence models, because the technology is changing so fast. Significant advances are expected throughout 2024 and beyond. But that’s precisely why we need to identify resources – to have a grounding for the next releases and updates.

When I started in Axapta, there was such a thing as a know-it-all, or seemed to be anyway. As the software grew (it is Enterprise Resource Planning (ERP) software after all), that idea went out the window and we became specialists.

Similarly with AI, we are at the ground level and we will benefit greatly from having a good overview of many of the big (popular) offerings out there.

Discussion

A burning question

AI Students’ frequently asked question: What are the tools people use most, according to Mark Thompson and Steve Little of The Family History AI Show (Episode 6). This is students who had a median of about 20 hours of experience in AI models prior to a weeklong set of 18 classes.

Well I’m not going to spoil, but you will have covered that and more in the prior four entries in this blog series.

Hands-on experience

Hands-on experience is considered crucial in learning AI and most topics. In all my Instructional Design courses and experience creating training courses at Microsoft, we strive to create the best learner experience which includes the hands-on work necessary to cement learnings and enhance problem-solving skills.

Hands-on experience is particularly crucial for AI due to several unique aspects of the field:

1. Complexity and Nuance: AI involves complex algorithms and models that require practical understanding. Hands-on experience allows individuals to see how these algorithms function in real-world scenarios, providing insights beyond theoretical learning.
2. Rapid Technological Advancements: AI is a fast-evolving field with new tools and technologies emerging regularly. Practical experience helps individuals stay updated with the latest developments and understand how to apply new tools effectively.
3. Interdisciplinary Nature: AI combines elements of computer science, mathematics, and domain-specific knowledge. Hands-on projects help integrate these diverse areas, enabling learners to develop a holistic understanding of how AI solutions are built and applied.
4. Problem-Solving Skills: Implementing AI solutions often involves tackling unstructured problems. Hands-on experience cultivates problem-solving skills, allowing individuals to experiment with different approaches and learn from failures.
5. Customization and Optimization: AI models often need to be tailored and optimized for specific tasks and datasets. Practical experience allows learners to understand the intricacies of model tuning and customization for different applications.
6. Ethical and Responsible AI: Working directly with AI tools provides insight into the ethical considerations and potential biases in AI models. This understanding is crucial for developing responsible AI systems.
7. Feedback and Iteration: AI development is iterative, involving continuous testing, feedback, and improvement. Hands-on experience facilitates this iterative process, enabling learners to refine models and approaches based on real data and outcomes.
8. Collaboration and Communication: Many AI projects require collaboration with other team members or stakeholders. Practical experience helps develop communication skills necessary for working effectively in multidisciplinary teams.

Overall, hands-on experience bridges the gap between theoretical knowledge and real-world application, making it essential for mastering the complexities and dynamic nature of AI.

Overview of available AI resources

(AI-generated spelling always cracks me up)

As in the rest of life, we will have to “learn to learn” so take the following list in that spirit, as inspiration and as a start, but not a complete static encapsulation.

• Please make sure to take a peek at a great “cheat sheet.” Steve [Little]’s Family History AI EDC [EveryDay-Carry] for Genealogists – oriented toward genealogists but broadly applicable. Note that he freely shares the information with a Creative Commons release. I may disagree with some of his categorizations (ahem!) but the best compact summary I’ve seen to date.
• Training: for free ones, see Paul Storm’s post of Denis Panjuta’s list here, and my blog entry 4 in 2. Take training of How to Adapt to AI-driven changes including my biased favorite Microsoft Learn; for paid see Coursera, edX, Udacity, Udemy, Pluralsight, and LinkedIn Learning.
• For news and updates, see Newsletters, blogs, podcasts, and the like. Some newsletters I get are The Rundown AI, Ethan Mollick from One Useful Thing, Charlie Guo from Artificial Ignorance, Eliot at Perplexity, Semafor Technology, and Last Week in AI.
• Community and collaboration – nothing beats learning alongside others. Search for communities including GitHub, which is a platform for sharing code.
• How do you get to Carnegie Hall? Practice, practice, practice! I’ve tried to give good and helpful examples in several earlier blog posts in this series. If you’d like to see more, drop me a comment! 

Challenge

Your hands-on work this week is to come up with a way that you can save a bit of time (or error) on something, and figure out how to make AI do this for you. Feel free to ask ChatGPT or another model for suggestions and help!

Summary

To coin a phrase, the best way to learn is to start. The beautiful thing about AI is that we don’t need to have mastery of it in order to see great benefits. Try the resources above and open up a whole new world.

To close, I’ll quote Professor Ethan Mollick in this LinkedIn post:

AI can’t do what you as an expert can do. But it can probably help you with something important that scatters your attention.

Disclosure

AI was used in several places in this post: to create the title, to create the outline, for ideas of trainings, for images.