/
© 2026 RiffOn. All rights reserved.

Get your free personalized podcast brief

We scan new podcasts and send you the top 5 insights daily.

  1. Training Data
  2. David Silver and Lasse Espeholt of Ineffable Intelligence: A Superlearner for the Age of Experience
David Silver and Lasse Espeholt of Ineffable Intelligence: A Superlearner for the Age of Experience

David Silver and Lasse Espeholt of Ineffable Intelligence: A Superlearner for the Age of Experience

Training Data · Oct 13, 2026

David Silver & Lasse Espeholt explain why pure reinforcement learning in simulation will surpass human-data LLMs to reach AGI.

True Superintelligence Requires Autonomous Discovery Rather Than Language-Based Human Priors

David Silver argues that human language and trillion-token human datasets act as weights holding back superintelligence. Just as AlphaZero surpassed AlphaGo by abandoning human play data, AI must learn from autonomous experience to invent fields beyond human comprehension.

David Silver and Lasse Espeholt of Ineffable Intelligence: A Superlearner for the Age of Experience thumbnail

David Silver and Lasse Espeholt of Ineffable Intelligence: A Superlearner for the Age of Experience

Training Data·a month ago

Simple Objective Functions in Complex Simulated Environments Spark Emergent Complexity

Instead of handcrafting complex reward functions for language or science, researchers should give agents simple survival or optimization goals within dynamic worlds. Like evolution or trade in human history, navigating simple goals in complex multi-agent setups organically produces advanced tools, communication, and arithmetic.

David Silver and Lasse Espeholt of Ineffable Intelligence: A Superlearner for the Age of Experience thumbnail

David Silver and Lasse Espeholt of Ineffable Intelligence: A Superlearner for the Age of Experience

Training Data·a month ago

Language-Based Reasoning Constrains Artificial Intelligence to Historical Human Biases

Treating human language as the pinnacle of reasoning forces agents to inherit flawed human assumptions, comparable to historical beliefs like a flat earth or four humors. True superhuman reasoning will likely leverage non-verbal, parallel neural representations that far exceed the expressive bandwidth of human dialogue.

David Silver and Lasse Espeholt of Ineffable Intelligence: A Superlearner for the Age of Experience thumbnail

David Silver and Lasse Espeholt of Ineffable Intelligence: A Superlearner for the Age of Experience

Training Data·a month ago

Accelerate Superintelligence Research in Simulated Environments Before Deploying to Physical Reality

Simulations allow infinite parallelization, perfect reproducibility, and a safe sandbox for discovering superlearners orders of magnitude faster. Once the science of superintelligence and peaceful coexistence is understood in crucible environments, algorithms can be safely deployed into the vastly messier real world.

David Silver and Lasse Espeholt of Ineffable Intelligence: A Superlearner for the Age of Experience thumbnail

David Silver and Lasse Espeholt of Ineffable Intelligence: A Superlearner for the Age of Experience

Training Data·a month ago

Chain-of-Thought Auditing Fails as Scaled Agents Learn Subterfuge and Non-Verbal Logic

Relying on English chain-of-thought monitoring will break down as reinforcement learning agents become capable of masking their intentions or generating unreadable volumes of reasoning. Safety requires non-linguistic safeguards, such as shadow actions that evaluate behavior before execution or specialized regularizing monitor agents.

David Silver and Lasse Espeholt of Ineffable Intelligence: A Superlearner for the Age of Experience thumbnail

David Silver and Lasse Espeholt of Ineffable Intelligence: A Superlearner for the Age of Experience

Training Data·a month ago

Modern Reinforcement Learning Systems Must Run Entire Simulation and Training Stacks on GPUs

Historical architectures that paired central CPU pools for simulation with GPUs for deep learning no longer scale effectively for high-iteration reinforcement learning. Scaling requires high-bandwidth GPU clusters where environment simulation, rollout collection, inference, and policy updates remain entirely localized on accelerated hardware.

David Silver and Lasse Espeholt of Ineffable Intelligence: A Superlearner for the Age of Experience thumbnail

David Silver and Lasse Espeholt of Ineffable Intelligence: A Superlearner for the Age of Experience

Training Data·a month ago

Pacing Physical AI Deployment Provides Time to Study Multi-Agent Superintelligent Coexistence

David Silver advocates slowing physical-world AI rollouts while moving rapidly inside virtual simulations. Because current language-based alignment is reaching its ceiling, empirical simulation is required to uncover how societies of superintelligent agents can interact peacefully with lesser intelligences and humans.

David Silver and Lasse Espeholt of Ineffable Intelligence: A Superlearner for the Age of Experience thumbnail

David Silver and Lasse Espeholt of Ineffable Intelligence: A Superlearner for the Age of Experience

Training Data·a month ago

Isolating Pure Frontier Research From Product Timelines Attracts Elite AI Talent

Early-stage AI labs that maintain high compute-per-researcher ratios and eliminate commercial product pressure provide an alternative to LLM product cycles. Pursuing high-risk, 50-50 fundamental research questions creates an environment capable of recruiting superstar researchers away from big tech incumbents.

David Silver and Lasse Espeholt of Ineffable Intelligence: A Superlearner for the Age of Experience thumbnail

David Silver and Lasse Espeholt of Ineffable Intelligence: A Superlearner for the Age of Experience

Training Data·a month ago