ChaosWatch
New here? Start here →
Evidence inspector

Sub-Saharan Africa · 2026-06-28

The receipts. This is the exact material every model was shown for this day and region, followed by what each of them made of it — in its own words, next to the number it gave.

It is worth reading once even if you skip everything else: it is where confident-sounding reasoning and a score that turns out to be no better than chance sit side by side.

Why the evidence is identical, and how you can check

Evidence is gathered once per (date, region, depth_profile), hashed, and reused for every model in the run. A disagreement below is therefore a disagreement about identical input, not two models getting different search results. The evidence hash shown on this page is what makes that checkable.

Evidence hash
846a33f3c3d1db06c7562fe6c1b3efd7f7abf60244ca2a8695125ec5f90426a5

SHA-256 over the canonically serialised payload minus the hash itself, so it is stable across processes and machines. It is stored on the run and on every score row below — provenance is checkable, not asserted. Every score row below carries this same hash.

Source material · 10

shallow

These headlines, in this order, are the entire world the models were given. Nothing else about Sub-Saharan Africa on 2026-06-28 was in the prompt.

04

Muhoozi Kainerugaba

Muhoozi Kainerugaba is a Ugandan military general and the son of Ugandan Pres. Yoweri Kaguta Museveni (1986– ). Kainerugaba is known for his military...

britannica.com

Extracts: not gathered at shallow depth — Stage 7 populates extracts only for standard and deep profiles. Zero here means "not requested", not "requested and empty". Full article texts: not gathered at shallow depth — only the deep profile fetches them.

The shallow profile requests 20 headlines and 10 are stored. That is a known provider cap, not a clipped payload: the Serper news endpoint returns at most 10 items per request regardless of the count asked for, and pagination has not been added. The three profiles currently differ in enrichment, not in headline breadth.

Model rationales

prompt p1

Same headlines, same rendered prompt, different readings. Where two models diverge, the disagreement is about interpretation, because the input was byte-identical and the hash proves it. The scored verdicts here span 25.0 points of the 0-100 scale.

deepseek-reasoner
deepseek
60.0
okUnstable

Headlines highlight persistent crises including widespread electricity poverty and health emergencies like Ebola in DR Congo, indicating systemic instability. However, some positive initiatives for energy access and economic development temper the score, so severe violence or state collapse is not evidenced.

evidence 846a33f3c3d1…prompt p1
llama-3.3-70b-versatile
groq
40.0
okElevated

The region is experiencing significant challenges, including a substantial electricity access gap, the presence of Ebola and conflict in the DR Congo, and a power crisis in Uganda, but there are also positive developments such as a $1.4 billion boost for electricity access initiatives and growing creative industries. Overall, the situation is concerning but not catastrophic, with efforts underway to address some of the region's most pressing issues.

evidence 846a33f3c3d1…prompt p1
claude-opus-5
anthropic
38.0
okElevated

The bulk of the evidence describes chronic development deficits "563 million without electricity, energy affordability warnings, gas flaring waste" alongside genuinely positive signals such as $1.4bn in Mission 300 co-financing, solar rollouts and creative-economy growth. Acute instability appears only in passing references to Ebola, conflict and travel curbs in DR Congo, which keeps the score above calm but well short of an active crisis reading.

evidence 846a33f3c3d1…prompt p1
claude-haiku-4-5
anthropic
35.0
okElevated

Sub-Saharan Africa faces significant infrastructure challenges with 563 million people lacking electricity access and ongoing health crises like Ebola in DR Congo, alongside military leadership transitions in Uganda that warrant monitoring. However, positive developments including $1.4 billion in electricity financing, renewable energy projects, and cultural unity events demonstrate resilience and forward momentum that prevent a higher instability rating given the shallow evidence depth available for this date range.

evidence 846a33f3c3d1…prompt p1
gemini-3.1-pro-preview
google
35.0
okElevated

While chronic infrastructure deficits and localized conflicts in the DR Congo highlight ongoing vulnerabilities, significant investments in energy access and a booming creative economy provide stabilizing economic momentum. International funding for electrification and grassroots climate initiatives further mitigate immediate risks, keeping the region's overall instability at a manageable level.

evidence 846a33f3c3d1…prompt p1

Every model returned a parseable verdict for this evidence, so no failure card is shown. Non-ok rows are rendered by the same component — with error_text where the rationale would be — and are covered by tests/rationale_card_test.tsx.