Source note: Local copy of a Summer Into AI 2026 submission originally published on Substack. Screenshots, video, and audio found in the post body are mirrored locally when publicly accessible; profile and avatar images are intentionally not copied. Read the canonical post.
Week 4 of Summer into AI 2026 hosted by @advisoryhour — theme: Built on Yesterday.
Earth Pulse watched the planet and narrated. Situation Room hands you the desk. Live USGS earthquakes, NOAA space weather, and city meteorology feed an AI watch officer that files a crisis briefing each round — real events, escalated into a developing scenario. You have ten response units and three rounds. Allocate them, execute, and the AI grades your call against the actual severities, then complicates the next round.
Worth noting alongside this: Eric's Black Budget Bird (Week 4) took AI-modeled geometry into the cockpit of an SR-71 flight sim. Situation Room stays on the ground floor of the war room — where the live data itself is the flight you cannot pause
Built on yesterday — what changed
The original: Earth Pulse (Week 3) — a passive monitoring station — it briefed you and you listened
You are now the decision-maker: ten response units, per-threat allocation, and consequences
The AI files a structured crisis briefing from the same live feeds (USGS, NOAA SWPC, Open-Meteo) — every threat on the board traces to a real event in the last 24 hours
A second AI judgment grades your allocation against the actual severities, awards a score, and generates the complication that reshapes the next round
A live global threat board plots every real earthquake of the last day by actual coordinates
How the AI works
Live-data briefing — three keyless public feeds are fetched fresh each round and handed to Claude, which files the briefing via a forced tool call: headline, situation, threats with severities, analyst recommendation. No two shifts are the same because the planet is not the same.
Consequence judgment — your allocation goes back to Claude with the threat data; a structured verdict returns per-threat outcomes from CONTAINED to CATASTROPHIC, casualties averted, a 0-100 score, and the next complication.
The complication loop — round two and three briefings receive both fresh live data and the previous complication, so your mistakes compound the story.
How to play
Press START SHIFT. The AI watch officer reads the live feeds and files your first briefing.
Study the threat board — severity bars are AI-assessed from real events. Allocate your ten units with the +/- controls.
Hit EXECUTE and face the after-action report: per-threat outcomes, casualties averted, and your score.
Three rounds, one shift score, and a final rank — from Distinguished Director down.
Where to play
Demo: week-4-demo-6-situation-room.vercel.app
Code & README: github.com/GlimmerForge/summer-into-ai → demo-06-situation-room/README.md
Summer into AI 2026 · Theme 4: Built on Yesterday
Original source
Canonical Substack URL: https://jakestrait5.substack.com/p/summer-into-ai-2026-situation-room.