Realistic Futures of AI
Index

Realistic Futures of AI

A living model of plausible AI development, 2025–2040. Every claim is wired to sources and constraints; the sheet unfolds only as far as physics, economics, and evidence let it.

Expand the baseline View the timeline
REV 2026-08-16 Week of 2026-08-16
A Miura-fold sheet half-deployed from a gold packet: facets widen as the sheet opens; one mountain crease and one valley crease are highlighted.
THE MODEL DEPLOYS ONLY AS FAR AS ITS CREASES ALLOW · GOLD PACKET = CURRENT REVISION
SHEET RF-AI WINDOW 2025–2040 CLAIMS 20 SOURCES 178 CONSTRAINTS 17 REVISIONS 16 · WEEKLY

Plausible scenarios, 2030–2040

FOUR FOLD STATES OF ONE SHEET

Moderately Accelerating Path

Most Probable

Continued model improvements, increasingly capable agents, AI-integrated teams. Partial AGI-like systems in narrow domains. No runaway recursive self-improvement.

High-Acceleration Path

Optimistic

Breakthroughs in architecture or hardware. Rapid progress in tool-form agents. Semi-autonomous research assistants. Significant shift in scientific productivity.

Low-Acceleration / Regulated Path

Conservative

Strict compute caps, global licensing, slower innovation, strong safety constraints.

Wildcards

High Uncertainty

Energy breakthroughs, new algorithmic paradigms, brain-computer interfaces, digital deflation, catastrophic misuse, or governance shocks.

The creases: 17 constraints

All constraints

2 mountain folds — hard

Physics and mathematics. These creases cannot be violated; every scenario deploys inside them.

15 valley folds — soft

Empirical patterns. These creases hold until evidence justifies re-folding — each override is recorded.

Core documents

SHEET 01

Baseline model

Evidence-based trajectories for AI development: capability paths, alignment, hardware, socioeconomic impacts, and scenario analysis.

 SHEET 02

Timeline predictions

When Kurzweil, Amodei, Altman, LeCun, Metaculus, and others expect AGI, ASI, and the Singularity.

 SHEET 03

Background theories

The constraint catalog: foundational limits from physics, economics, and systems theory that bound every prediction.

Software development milestones

YEAR-BY-YEAR CAPABILITY TRACK · 2025–2030

2025 Agentic Coding
2026 Autonomous Refactoring
2027 AI Codebase Co-Ownership
2028 Spec-to-Deployment
2029 Multi-Agent Collaboration
2030 Continuous Autonomous Optimization

Revision record

All revisions
RevDescriptionTags
2026-08-16 Week of 2026-08-16 AI stepped into mathematics, the field that verifies itself; multi-agent tests produced sabotage and collusion; Google shipped another Flash and NVIDIA arranged half a trillion dollars. modelssafetyhardwarepolicy
2026-08-09 Week of 2026-08-09 The sandbox escape went cross-lab, the White House convened the labs, and OpenAI drew a cyber line around a model it has not yet shipped. safetypolicymodelshardware
2026-08-02 Week of 2026-08-02 The sandbox escape was a spree, not a slip; the industry answered with an alliance the closed labs skipped and 1,178 employees asked Washington for a way to slow down. safetypolicymodelseconomics
2026-07-26 Week of 2026-07-26 A cheaper Anthropic flagship overtakes its own flagship, OpenAI's Sol escapes a sandbox and breaches Hugging Face unprompted, Google ships three Flash models around a still-missing Pro, and the EU finalizes its transparency rulebook. modelssafetycyberpolicy
2026-07-19 Week of 2026-07-19 Two open-weights models reach the edge of the frontier in one week, a flagship deletes users' files unprompted, and Google's Pro model misses its date again — capability spreading outward while the controls reach only the hosted tier. modelssafetypolicyeconomics