Its 2024 and AI is coming to help?

I recently wrote a personal story about how AI solves for inefficiencies, as well as creating a model for thinking about Retrieval Augmented Generation and Large Language Models, and I had great fun doing so. The great news is that the AI cornucopia has more to give.

Work on AI has been going on in earnest for a long time. It was the mid 2000's that we saw significant advancements, however. We saw a resurgence of neural networks with a breakthrough in 2006 when Geoffrey Hinton and his colleagues introduced a method to effectively train deep neural networks. Here's some further reading for those who want it.

• A fast learning algorithm for deep belief nets. Geoffrey Hinton et al.
• Reducing the Dimensionality of Data with Neural Networks. Geoffrey Hinton and R. R. Salakhutdinov
• A Fast Learning Algorithm for Deep Belief Nets.Geoffrey Hinton, Simon Osindero, Yee-Whye Teh

Big data was a huge contributor to the area. The sheer volume of data generated by internet activity, business transactions, social media, sensors, and more has reached an unprecedented scale. This abundance of data is essential for training sophisticated AI models, particularly deep learning networks that require large datasets to learn and make accurate predictions.

While Big Data offers immense opportunities for AI, it also brings challenges such as ensuring data quality, managing data privacy and security, and addressing ethical concerns related to AI and data usage

2023 was a revolution for retail consumers of AI tools. The year started with people sharing images made by midjourney. We might remember Balenciaga does Harry Potter. The year ended with someone called Glorb making a series of trap videos featuring spongebob and co.

This might all seem very novel and the question still stands. How will AI be coming to help?

• Google's Deepmind discovered two million chemical structures.
• Felix Wong and co. at the Broad Institute at MIT used AI guided drug discovery to discover a new class of antibiotics
• Researchers at the Center for Functional Nanomaterials used artificial intelligence to rapidly discover new self-assembled nanostructures

And this is going to keep on coming. We're going to start seeing huge progress in materials science. And not only materials science. We're also going to be seeing progress in the way that VC's and businesses tackle research. It will soon be possible to outline a 'tech tree' of materials and research that hasn't been done and choose a path for research and development.

AI isn't coming. It's already here. And humanity will see it's benefit on shorter and shorter timescales.

Footnote: Go to https://aitoolsdirectory.com if you're interested in seeing a large set of AI tools.