Rafael’s Substack about AI
The explosive popularity of chatbots and generative AI models has introduced the concept of artificial general intelligence (AGI) to society faster than ever before. Many experts and enthusiasts now believe that AGI—software capable of performing any intellectual task that a human can—is not only possible but also imminent. Enough so to move markets, scramble governments, and rekindle interest in the technological singularity and other transhumanist ideas. Amid growing cracks, from the diminishing returns of scaling laws to unclear profitability and dissenting voices, the hype is set to continue. Between the sudden factual advances of AI capabilities, heralded by transformer-based Large Language Models, unashamed predictions from stakeholders and fearmongering, notably by those that for decades taught and inspired many of today’s AI researchers, 2025 promises to be another year of surprises that will test the limits of Generative AI. Whether the data-hungry imitators that constitute the latest generation of generative models are sufficient for AGI is a hot topic of debate.
Yet, because it’s holiday season and we’re all recouping from another eventful year, let’s table this question for another post to engage instead in a lighthearted discussion about the coming (fictional) “technological immortality” envisioned by transhumanists to close off 2024.
