▶ ai·
After Orthogonality: Virtue-Ethical Agency and AI Alignment
Peli Grietzer on eudaimonic rationality and its implications for AI alignment.
items20 latest
▶ ai·
2025 in AI, with Nathan Benaich
AI in 2025, going into 2026
▶ ai·
Iason Gabriel: Value Alignment and the Ethics of Advanced AI Systems
How do we develop AI systems reflective of human values? How do we determine those values?
▶ ai·
AGI is Not Multimodal
Benjamin Spiegel on what disembodied definitions of AGI miss
▶ ai·
2024 in AI, with Nathan Benaich
all the (AI) news that's fit to print (record).
▶ ai·
Philip Goff: Panpsychism as a Theory of Consciousness
On phenomenal transparency, physicalism vs. anti-physicalism, hybrid cosmopsychism and its assumptions
▶ ai·
Some Changes at The Gradient
How we're scaling back a bit to keep things sustainable
▶ ai·
Shape, Symmetries, and Structure: The Changing Role of Mathematics in Machine Learning Research
Henry Kvinge asks "What is the Role of Mathematics in Modern Machine Learning?"
▶ ai·
Jacob Andreas: Language, Grounding, and World Models
On grounding, language and the world, agents, and world models.
▶ ai·
Evan Ratliff: Our Future with Voice Agents
On things that are not what they seem; and reflections on the Longform Podcast.
▶ ai·
Meredith Ringel Morris: Generative AI's HCI Moment
On distinctions between fields, technological determinism, disability and AI, and why generative AI research needs more humanism.
▶ ai·
Update #83: AI Music Fraud and PlanSearch
We look at the mechanics of AI-assisted music streaming fraud; researchers develop a new algorithm for LLM code generation that leverages inference time search over high-level plans
▶ ai·
Mini-Update #47: First International AI Safety Treaty and the WavTokenizer Codec
The UK signs on to the first international AI safety treaty and a novel audio codec model allows for high-quality information transmission in fewer tokens.
▶ ai·
What's Missing From LLM Chatbots: A Sense of Purpose
Kenneth Li argues that LLM chatbots are missing a "goal" for each conversation and proposes fixing this with human-in-the-loop model evaluations.
▶ ai·
Davidad Dalrymple: Towards Provably Safe AI
On provably safe and beneficial AI and the UK's new research funding agency.
▶ ai·
Clive Thompson: Tales of Technology
On writing about the climate, theories of change, the anthropology of technologists, and where AI might be going.
▶ ai·
Update #82: AI Lawsuits and SOPHON
We look at the numerous ongoing lawsuits against AI giants; researchers develop a method to address the risk of pre-trained models being repurposed for unethical or harmful tasks.
▶ ai·
Judy Fan: Reverse Engineering the Human Cognitive Toolkit
On how we use physical representations of thought to learn and communicate, cognitive tools' benefits and risks, and science as process.
▶ ai·
Mini-Update #46: OpenAI-Condé Nast Partnership and Flexible Perception
Major publishing company Condé Nast grants OpenAI the rights to train models on their work, and Google Deepmind researchers integrate visual memory retention to augment model flexibility.
▶ ai·
L.M. Sacasas: The Questions Concerning Technology
On contingency, our narratives about and engagement with technology, and what it is to be human.