Insights for Future Artificial Intelligence Developers

Artificial Intelligence (AI) is not a novel concept and has been a staple in academic and technological sectors for decades, studied by professionals who have contributed to its development, and a topic of intense discussion at the Davos Forum last year.

Yet, over the past year, there's been a significant shift. The term "Artificial Intelligence" now pervades various industry talks, with businesses and governments associating themselves with the subject along the Promenade at the event, seeking to engage their corporate audience or position themselves at the forefront of this field. This is a positive development, though it can also lead to anxiety, confusion, and information overload for those looking to get involved, incorporate AI into their professional realm or business, or specialize and pursue a career in this fascinating field.

[ I plan to share more of my notes and thoughts on this broad topic on LinkedIn, particularly focusing on the “what” and “how” of using AI more effectively, improving work across different fields, and for technologists and engineers, learning how it functions. ]

Regardless, I believe it's vital to recognize the difference between the demands of businesses, technology professionals, and the general population's use of this tool. Clearer communication is essential to make AI accessible and beneficial for both experts and the general public.

I want to start by thinking about those interested in working behind the scenes of AI. (I also invite friends in the field to add their comments below, enriching this article with their perspectives and personal experiences.)

After what I've seen and discussed here, my first recommendation for professionals wanting to work in AI is to be wary of shortcuts, magic formulas, or new gurus. Ensuring a solid foundation and understanding of essential subjects like Statistics, Mathematics, R, Python, Machine Learning, neural networks, and development infrastructure, among others, is crucial. These are the pillars of knowledge that I often see in conversations with experts, helping students or professionals understand, contribute to, and create AI models that benefit various businesses and society.

Another great source of free learning is the history of many organizations and individuals pioneering in AI. For instance, through a ChatGPT product or the story of OpenAI , one learns how to make technology relevant to people, which has sparked a true revolution. Similarly, it's important to learn from the trajectory and product development of companies like IBM , with its IBM watsonx technology, which not only made what we know of Machine Learning and AI today possible but continues to drive AI use and evolution by businesses and governments. Understanding the history and contributions of these agents is inspiring for those who want to stand out in the field.

As a third suggestion, it's crucial to know about the many brilliant minds that have enabled this evolution. For example, Yann LeCun , now the chief AI scientist at Meta, whose journey and impact are internationally recognized and awarded (https://www.nytimes.com/2019/03/27/technology/turing-award-ai.html). A genius I had the privilege to meet and learn from during my time at Meta, and whom I always follow, as decoding his trajectory, projects, and contributions are valuable knowledge sources for anyone interested in being part of this revolution. His papers are rich, and there are numerous videos, talks, and classes of his available online. For reading, I recommend the 2015 "Deep Learning" article published in Nature, where he and his co-authors provide a comprehensive overview of the deep learning field, summarizing decades of research and developments: https://www.nature.com/articles/nature14539

To conclude, I know there's a vast array of academic literature and online resources, but I mention here some references and resources, both familiar and recurrent in conversations this week, useful for those already in the field or just starting:

1. “Artificial Intelligence: A Modern Approach” by Stuart Russell and Peter Norvig - Commonly used in introductory AI courses, this book covers a wide range of topics in artificial intelligence.

2. "Pattern Recognition and Machine Learning" by Christopher M. Bishop: An essential resource to understand machine learning concepts.

3. "Deep Learning" by Ian Goodfellow, Yoshua Bengio , and Aaron Courville: A comprehensive introduction to deep learning.

4. Online courses offered by Coursera or edX, focused on Python, Machine Learning, and AI.

5. Following recent publications on Google Scholar and participating in forums like Stack Overflow and GitHub to stay updated and engaged in the community.

Finally, I am thrilled to experience and participate in such a moment of transformation and possibilities that technology is allowing us to create now and in the future. The ongoing changes are enormous, with an impact greater than the revolution brought by access to PCs and later by the Internet. We are now witnessing the dawn of a new era that, despite its challenges and the need for protective measures (also intensely discussed at the Forum), I believe opens up a range of opportunities, with the potential to promote inclusion and prosperity like we've never seen before for all nations.

Written by Daniel Arantes (Threads in Portuguese: @danielarantes )