What are some seminal moments for AI in the last 5 years that made it everything, everywhere, all at once?

Artificial intelligence (AI) is everywhere. From our smartphones to our cars, from our social media feeds to our online shopping, AI is shaping our lives in ways we may not even notice. But how did we get here? What are some of the seminal moments for AI in the last five years that made it everything, everywhere, all at once?

In this blog post, I will highlight some of the most important advances and achievements in AI from 2019 to 2023, and how they have impacted various domains and industries. This is not a comprehensive list, but rather a selection of some of the most notable and influential milestones that have pushed the boundaries of AI research and applications.

- 2019: OpenAI releases GPT-2, a large-scale language model that can generate coherent and diverse texts on various topics. GPT-2 is a breakthrough in natural language processing (NLP), as it demonstrates the power of deep learning and self-attention to capture the nuances and complexities of natural language. GPT-2 can generate realistic texts on almost any topic, given a few words or sentences as input. It can also answer questions, summarize texts, write essays, and even create fictional stories. GPT-2 is both impressive and controversial, as it raises ethical concerns about the potential misuse and abuse of such a powerful technology.

- 2020: DeepMind's AlphaFold solves the protein folding problem, a longstanding challenge in biology and medicine. AlphaFold is a deep learning system that can predict the three-dimensional structure of proteins from their amino acid sequences, with unprecedented accuracy and speed. Protein folding is crucial for understanding how proteins function and interact in living organisms, and how they can be modified or designed for various purposes. AlphaFold's achievement is a milestone for computational biology and biotechnology, as it opens up new possibilities for drug discovery, disease diagnosis, and synthetic biology.

- 2021: Tesla launches its Full Self-Driving (FSD) beta software, enabling its electric vehicles to navigate complex urban environments autonomously. FSD is a suite of advanced driver assistance systems (ADAS) that use computer vision, deep learning, and sensor fusion to perform various driving tasks, such as lane keeping, lane changing, parking, traffic light detection, and obstacle avoidance. FSD is one of the most ambitious and challenging applications of AI in the automotive industry, as it aims to achieve level 5 autonomy, where the vehicle can drive itself without human intervention in any scenario.

- 2022: OpenAI releases DALL-E, a generative model that can create images from text descriptions. DALL-E is a combination of GPT-3, a larger and more powerful version of GPT-2, and CLIP, a vision model that can learn from natural language supervision. DALL-E can produce realistic and diverse images from any text input, such as "a cat wearing a hat" or "a painting of a snail made of harp". DALL-E can also manipulate images based on text instructions, such as "the same cat without a hat" or "the same snail but bigger". DALL-E demonstrates the ability of AI to bridge the gap between language and vision, and to generate novel and creative content.

- 2023: DeepMind's MuZero achieves superhuman performance in multiple domains, including chess, Go, shogi, Atari games, and StarCraft II. MuZero is a reinforcement learning algorithm that can learn to master complex environments without any prior knowledge or rules. MuZero uses a neural network to model the dynamics and rewards of the environment, and to plan its actions accordingly. MuZero can adapt to different tasks and challenges, and can even invent new strategies and tactics that surpass human experts. MuZero is a breakthrough for artificial general intelligence (AGI), as it shows the potential of AI to learn from its own experience and to excel in diverse domains.

These are just some of the examples of how AI has advanced and transformed in the last five years. AI is not only a scientific endeavor, but also a social and cultural phenomenon that affects every aspect of our society. As AI becomes more ubiquitous and powerful, we also need to consider its ethical implications and societal impacts. How can we ensure that AI is aligned with our values and goals? How can we foster trust and collaboration between humans and machines? How can we leverage AI for good and avoid its pitfalls? These are some of the questions that we need to address as we enter the next era of AI.