← Originals
Greg Brockman

Greg Brockman

6 interviews

What they argue

Greg Brockman is a thinker focused on the architectural philosophy and operational shifts of artificial intelligence. He explores the transition from static, offline model training to dynamic reasoning and the safety implications of deploying AGI-scale systems.

  1. The Reasoning Paradigm

    The current frontier of AI is a shift from mere next-token prediction to a model's ability to navigate complex, multi-step logic. This distinguishes a simple chat bot from an agent capable of solving unsolved problems.

    From Greg Brockman on OpenAI's Road to AGI (episode)

    • Says much the same Nathan Labenz — The Shift from Prediction to Task CompletionModern AI models are moving beyond simple next-token prediction toward being trained to achieve specific goals via reinforcement learning (RL). This shift means models…AI Scouting Report: the Good, Bad, & Weird @ the Law & AI Certificate Program, by LexLab, UC Law SF
    • Says much the same Henry Kissinger — The Shift from Tools to AgentsUnlike traditional technologies that act as extensions of human intent, AI is an agentic technology that can generate novel solutions and decisions. This represents a…The Age of AI

  2. The Reasoning Loop Framework

    To move a model from knowledge to reasoning, one should follow a loop of hypothesis generation and environmental interaction. This allows models to test hypotheses and receive feedback from reality.

    From Greg Brockman on OpenAI's Road to AGI (episode)

    • A related idea Gary Klein — The Mental Simulation LoopTo build intuition, one must move from 'Reaction' to 'Simulation' by identifying patterns and 'chunks' of past experience. This process involves evaluating potential…The Power of Intuition
    • A related idea Jeff Hawkins — The Predictive Processing LoopTo understand how an intelligent agent interacts with its environment, one can use a loop of generation, comparison, and correction. The system generates a 'prior'…On intelligence

  3. The AGI Era as a Phase

    AGI is not a single switch that flips but a fuzzy spectrum. The current era is defined by models that can run coherently for long periods, such as 24 hours, to accomplish complex tasks.

    From OpenAI President on What it Means to Cross Into The Age of AGI (episode)

  4. Shift to Safety and Alignment

    The transition into the AGI era marks a shift from a 'compute-first' race to a 'safety-and-alignment-first' era. The primary challenge is ensuring that increasingly capable models are distributed safely and benefit everyone.

    From Greg Brockman on What it Means to Cross Into The Age of AGI (episode)

    • Says much the same Simon Willison — The Slow Takeoff PhilosophyAGI should be deployed incrementally to manage safety risks and ensure controlled integration into the global economy. This explains why certain features, such as full…AGI is Being Achieved Incrementally (OpenAI DevDay w/ Simon Willison, Alex Volkov, Jim Fan, Raza Habib, Shreya Rajpal, Rahul Ligma, et al)
    • A related idea Ilya Sutskever — The Transition to the Research EraThe 'Age of Scaling' is reaching its limits because more data and compute do not inherently guarantee the emergence of taste or judgment. The next era of AI development…Ilya Sutskever – We're moving from the age of scaling to the age of research

  5. The Defense Factory Framework

    To survive AI-driven cyber threats, organizations must move from static security to a machine-speed 'Defense Factory.' This involves using frontier models to identify vulnerabilities in existing code and infrastructure.

    From OpenAI President on What it Means to Cross Into The Age of AGI (episode)

Every idea names the summary it came from. Written by a local model from this mind’s own work, never from memory.

Interviews 3

Open Greg Brockman in Originals