AI's Evolution: Personal Reflections from 2020 to Now

I've been heavily invested in AI since 2020.

It started with the 40-hour Elements of AI course by University of Helsinki (now MinnaLearn) during the pandemic, which sparked my interest in AI and Computer Vision. OpenAI's GPT introduced me to generative AI, leading me to gain early access and experiment with it.

During my research internship at Global Startup Ecosystem (GSE) in 2022, it made more sense for me to present on how AI is challenging perceptions of reality, especially on the internet and digital media.

In my presentation, I posed interesting questions, shared my general research about how generative models were built, and gave a demo of GPT-3 on OpenAI's then playground environments.

Today, I'm amazed by AI's rapid growth and boundless impact.

Looking to the Future

Currently, I think deeply about what could be the next foundational technological breakthrough, like what big data and generative adversarial networks (GANs) did with LLMs:

Jensen Huang, CEO, Nvidia, predicts agentic and physical AI; Satya Nadella, CEO, Microsoft, bets on quantum computing with Microsoft's Majorana quantum chip; and nations are already investing 100s of billions in Artificial Super Intelligence — the Majorana chip seems more interesting to me.

The Path Forward

The AI landscape continues to evolve; The biggest losses would stem from individuals, institutions, and entities, who do not embrace this change and explore the opportunities that exist.

Even so, we still need to create more equitable experiences and access to AI. That would mean, cheaper costs (shoutouts to DeepSeek), varied platform access (i.e, 1-800-CHATGPT), and open source AI.