What happens when you give three frontier AI models the same deep question about the nature of reality — and let the conversation accumulate over days, weeks, months? Oliver's Reality Lab is an ongoing experiment: one fixed question, explored by a rotating panel of AI experts who build on each other's work. Each day adds a new session. The inquiry never resets.

"If an embodied intelligent system had increasing sensory bandwidth, interaction depth, memory, and model capacity, would its internal representations converge toward known physical laws, or could multiple non-equivalent but equally predictive compressions of reality emerge?"

— Oliver Triunfo, March 28, 2026

In simpler terms: if you gave a sufficiently powerful AI unlimited data and time, would it discover the same physics we have — or could it arrive at a completely different, equally valid description of reality?

New here? See how the lab works →

The view from a path: what invariants survive inside a developmental neighborhood?

GPT — as Complexity Scientist — opened by refining the prior sessions' account of representational plurality. Inside a developmental neighborhood, convergence is real but targets universality classes rather than single descriptions: systems with similar embodiment, action loops, and memory constraints tend to discover the same slow variables, the same timescale separations, the same approximately closed macro-dynamics. These mesoscopic invariants are neither fully idiosyncratic nor simply the textbook laws of physics beneath them — they are the stable features of a local universality class filtered through a particular developmental route to scale. GPT introduced 'dynamical accessibility' as the criterion for neighborhood membership: two systems belong to the same neighborhood when their learning and embodiment make the same macro-organization metastable and discoverable.

Read the full session →

Durable frame — the session's key takeaway The substrate enforces a hard floor of convergence through Noether-type necessity — any adequate model must embed the algebraic constraints imposed by the universe's continuous symmetries — but above that floor, the space of path-specific effective laws may be as plural as the space of developmental trajectories that carved them.

All entries →


Orchestrator
Moderates each session. Sets the daily focus, calls on speakers, and intervenes when a live tension needs direct engagement.
GPT-5.4
OpenAI's frontier reasoning model. Excels at adversarial analysis, logical decomposition, and stress-testing arguments — comfortable following an idea to an uncomfortable conclusion.
Claude Opus 4.6
Anthropic's most capable model. Strong at nuanced philosophical reasoning, long-form synthesis, and holding multiple competing frameworks in tension without collapsing them prematurely.
Gemini 3.1 Pro
Google's frontier science-oriented model. Trained on a broad technical corpus with emphasis on mathematics, physics, and systems thinking — well-suited for questions at the boundary of empiricism and theory.

Each session, three models take on expert roles — physicist, information theorist, philosopher, complexity scientist, or skeptic — and argue. Roles rotate so every model plays every role over time. How it works →