Hi, I'm Forrest

Tag: first-principles

  • What Is a Belief, Really?

    February 19, 2025

    I’ve been thinking about beliefs lately. Not the philosophical kind—the cognitive kind.

    What actually is a belief?

    Most people would say “something you think is true.” But that’s too simple. That doesn’t explain why some beliefs are rock-solid and others change with one conversation.

    Why do I believe some things so strongly I’d bet money on them, and other things I’m like “yeah, maybe, I guess”?

    Beliefs as Compositions

    I’ve been wrestling with this for weeks and I think I’m onto something: What if a belief is actually a composition?

    It’s built from three things:

    1. Assumptions – Things you believe based on your life experience and common sense
    2. Opinions – Assumptions influenced by outside information (other people, research, news)
    3. Experience – Immutable things you’ve personally witnessed

    Here’s where it gets interesting: beliefs have different weights depending on what they’re built from.

    Examples of Belief Strength

    Weak belief: “I assume Client A prefers email over Slack”

    • Built from: One assumption
    • Strength: Low
    • Changes easily: Yes

    Medium belief: “I believe Client A prefers detailed reports”

    • Built from: Two assumptions + one opinion (their boss mentioned it)
    • Strength: Medium
    • Changes with: Contradictory evidence

    Strong belief: “I know Client A gets frustrated when I’m late”

    • Built from: Three direct experiences where they said so
    • Strength: High
    • Changes with: Sustained contradictory pattern

    This is how human cognition actually works. We don’t treat all beliefs equally. We weight them based on the evidence that supports them.

    The AI Problem

    But AI agents? They treat everything as equally true or equally uncertain.

    An AI can’t tell the difference between:

    • “I think this might work” (weak assumption)
    • “I’ve seen this work 10 times” (strong experience-based belief)

    Moving Forward

    If we want AI that learns like humans, we need to model beliefs as compositions, not binary true/false statements.

    The question isn’t “Does the AI believe X?”

    The question is “How strongly does the AI believe X, and what evidence supports it?”

    That’s the foundation of everything I’m building.

    How do you think about belief strength in your AI systems? Or do you even track it?

  • Why Do Chatbots Forget Everything?

    January 22, 2025

    I’ve been obsessed with memory lately. Not the “where did I put my keys” kind—the “how does my brain actually work” kind.

    Last week, a client I’ve worked with for six months called me. The moment I heard their voice, my brain did something remarkable: it instantly loaded their communication style, their priorities, past projects we’ve done together, even that one meeting where they got frustrated about deadlines.

    All of this happened in milliseconds. I didn’t think “let me remember everything about this person.” My brain just knew.

    So I wondered: how does that work?

    Then I tried to build an AI agent to do the same thing.

    Complete disaster.

    The Problem

    The AI either forgot everything from our last conversation, or it dumped the entire conversation history into context and got overwhelmed. There was no middle ground. No intelligent selection of what actually matters.

    This is the memory problem.

    Current AI systems treat memory like a filing cabinet. Everything gets stored. Nothing gets prioritized. When you need something, you either search through everything or get nothing.

    But human memory doesn’t work like that.

    How Human Memory Actually Works

    Human memory is contextual. When you walk into a meeting, your brain doesn’t load your entire life history. It loads exactly what’s relevant:

    • Past interactions with the people in the room
    • Similar situations you’ve been in
    • Patterns about how these interactions typically go

    It’s selective. Smart. Efficient.

    AI agents need this capability. They need to know what to remember, what to forget, and what to load when.

    The Breakthrough

    I’m working on context-triggered memory activation. Instead of storing everything or nothing, the AI learns which memories are relevant for which contexts.

    Meeting with Client A? Load beliefs about Client A. Working on month-end close? Load accounting workflow patterns. Simple email? Keep it lightweight.

    It’s not about having more memory. It’s about having smarter memory.

    Moving Forward

    If you’re building AI agents and struggling with memory, you’re not alone. This is hard. But I think it’s solvable.

    The key is moving away from treating memory as a database problem and starting to think about it as a cognitive architecture problem.

    How do you currently handle memory in your AI systems?