Proficiency in using AI involves clear communication, skepticism, and a deep understanding of both the technology and the domain in which it is applied.
About:
Better playing through algorithms
Website:
Specializations:
Incoming Links:
Subscribe to RSS:
AI's rise necessitates political discourse on power concentration, responsibility, intellectual property, and labor displacement, advocating for open-source policies and democratic engagement.
The author argues against the automation of science, emphasizing the importance of human involvement and the meaningfulness of contributing to research.
AI-generated papers threaten scientific integrity, necessitating a reformed peer review process that emphasizes accountability and community engagement.
5.7AI Allergy
2025-08-13 • technology innovation agi critique ai
The author expresses frustration with the current state of AI discourse, feeling overwhelmed by hype and unrealistic expectations surrounding AI advancements, particularly regarding AGI and the singularity. They reflect on the his...
Google DeepMind has introduced Genie 3, an advanced neural game engine that generates interactive environments based on user prompts. The model excels in creating photorealistic settings but has limitations, including control inpu...
A successful AI researcher reveals their lifelong struggle with math, arguing that proficiency in mathematics is not a prerequisite for success in computer science.
The post presents a thought experiment on the implications of requiring all sufficiently expensive machine learning models (costing a million dollars or more) to be open-sourced. It argues that this would eliminate the competitive...
The post compares the science fiction universes of Star Trek and The Culture, highlighting their similarities and differences, particularly in their portrayal of advanced technology, societal structures, and the role of artificial...
The author reflects on the potential for automation to replace various jobs, such as doormen, maintenance workers, teachers, and police officers. They also discuss the nature of their own job as a faculty member and the potential ...
The text describes the concept of language and symbols coming alive in a library, reproducing themselves without human intervention, and the implications of this phenomenon on the value and essence of language.
The author discusses the death of the lecture as a mode of knowledge transmission, reflecting on the history of lectures, the interactive nature of reading, and the impact of online courses and AI on lectures.
The author, Togelius, is well-known but not well-understood. He asks new LLMs to write an essay about him, and most of them get it wrong. This exercise helps him understand the inaccuracies and the concept of Gell-Mann amnesia.
The text discusses the evolution of smartphones and compares it to the current state of large language models. It speculates on the future of large language models and their impact on society.
The author argues that the term AGI is lazy and misleading, as it is undefinable and unlikely to ever be a useful concept. He explains that the term is vague and ambiguous, leading to bad discussion, limited understanding, and ult...
0On the "economic definition" of AGI
2024-09-26 •
The text discusses the economic definition of AGI and how it is a moving goalpost phenomenon. It argues that humans will always find new economically valuable work to do, and that the technology hyped right now is mostly good at s...
0Artificial General Intelligence (the book) is here!
2024-09-24 •
The author's book on Artificial General Intelligence has been officially released, and it aims to make the topic accessible to a wide audience. The book discusses the confusion surrounding the idea of artificial general intelligen...
0AI safety regulation threatens our digital freedoms
2023-11-01 •
The blog post discusses the potential threat of AI to humanity and the need for regulation. The author argues that regulating AI development in the name of AI safety would require unprecedentedly invasive surveillance of personal ...
0As A Large Language Model, I
2023-09-17 •
The text is written from the perspective of a large language model, discussing its limitations and capabilities. It reflects on the model's inability to think backward, its role in culture, and its ability to take on different rol...
The blog post discusses the concept of intelligence and how it is perceived. It explores the idea of intelligence as a scalar quantity and presents various analogies to help understand it. The author argues that the view of intell...
0Is Elden Ring an existential risk to humanity?
2023-04-04 •
The blog post discusses the existential risk from superintelligent AI and compares it to the existential threat posed by the video game Elden Ring. The author argues that the concept of 'intelligence' is ambiguous and that the doo...
0The Cult of Gai
2022-11-30 •
The blog post discusses a fictional cult that believes in the appearance of a deity named Gai, who will destroy humanity. The cult is obsessed with trying to stop this from happening, but their beliefs are based on technology and ...
0Apology for Video Games Research
2022-08-08 •
The blog post discusses the history of video games research and the importance of video games in AI research. The author talks about the skepticism and hostility faced by researchers in the early days of digital computing and how ...
0Brief statement of research vision
2022-07-29 •
The author discusses his research vision of creating more general artificial intelligence through training sets of diverse-but-related agents in persistent generative virtual worlds. He emphasizes the need for diversity and shared...