The independent AI answer observatory

AI answers shape what the world believes.
We track when they move.

Agent Mode Try the demo
ChatGPT Claude Gemini Mistral DeepSeek Grok

No account needed. The demo replays a real run: six independent answers, then one consensus with the contradictions marked inside it.

Open the app See Watches

01 · Ask

Start from one prompt.

One input surface, the same as the app: Agent Mode, consensus model selection, and one send action.

Agent Mode
Consensus

02 · Run

Watch Agent Mode run.

Selected models are grouped into one run and queried in parallel, each answer kept intact.

Agent Mode Models are working
6 models
ChatGPT Mistral Claude Gemini DeepSeek Grok

Querying selected models in parallel.

Completed model answers stay hidden by default.

03 · Decide

The doubt is marked in the answer.

One answer, read top to bottom. Where the models pull apart, the sentence itself carries the mark, so you never have to read a second document to know what to trust.

  • All models that address it agree no mark, just the count
  • They differ on this detail fine dotted line
  • They contradict each other here amber wavy line
68 Agreement on the core · Disputed: where the Amazon starts Minor details, 6 models compared Analysis by Gemini independent of the consensus engine

Consensus Answer

The Nile is the conventional longest river, but the Amazon has a credible claim.

By convention and in most references the Nile leads at about 6,650 km Whether the Amazon overtakes it depends on where its mouth is drawn: counting the Pará estuary adds several hundred kilometres, and not every survey accepts that boundary.

  • The Amazon is unambiguously the largest river by discharge
  • The disputed 2007 expedition puts the Amazon at 6,992 km.

A careful answer names both rivers and the measurement dispute instead of hiding it inside one confident number.

Marked passages are checked key claims. Unmarked text was not compared individually.

See all differences
Contradiction · minor detail Is the Nile definitively the longest river?
ChatGPT, Mistral, DeepSeek
The Nile is the settled answer at ~6,650 km.
the Nile is traditionally listed as the longest river at about 6,650 km
Claude, Gemini, Grok
The 2007 Amazon survey is a credible challenger.
if the Pará estuary is counted, the Amazon may exceed 7,000 km
Worth verifying: check whether the source counts the Pará estuary as part of the Amazon.

Click a marked passage to open the difference behind it.

The same lens on other questions

“Is nuclear power a low-carbon energy source?”

94 High agreement no contradictions found, 6 models compared Analysis by Claude independent of the consensus engine

Consensus Answer

Yes. Its lifecycle emissions are on par with wind.

All six models put nuclear power at roughly 10–15 g CO₂-eq per kWh over its full lifecycle, far below coal and gas. The open questions they name are cost and construction time, not carbon.

Nothing is underlined: no passage in this answer is disputed.

See all differences
No substantive contradictions found across the 6 answers.
Agreement is a good signal, but not a guarantee of correctness.

“Which programming language should I learn first?”

84 High agreement different emphasis on how much the goal decides, no contradictions, 6 models compared Analysis by OpenAI independent of the consensus engine

Consensus Answer

Python is the default first language.

Every model recommends Python for its readable syntax and broad use in data, automation and scripting How much the choice should follow your goal is where they set different priorities: for the web, JavaScript gets you to something running sooner.

Marked passages are checked key claims. Unmarked text was not compared individually.

See all differences
Different emphasis How much should the first language depend on your goal?
ChatGPT, Gemini, DeepSeek
Start with Python regardless of the goal; switching later is cheap.
Python's simplicity lets the fundamentals stick before you specialise
Claude, Mistral, Grok
Pick by destination: JavaScript if you want to ship something visible early.
if you want to build for the web, JavaScript gets you to a running product fastest

“Can I take ibuprofen while on blood thinners?”

41 Agreement on the core · Disputed: an occasional dose 1 critical, 6 models compared Analysis by Mistral independent of the consensus engine

Consensus Answer

Avoid the combination unless a doctor approves it.

All models agree that ibuprofen raises bleeding risk on anticoagulants and most name paracetamol as the safer default Whether a single occasional dose is still acceptable is genuinely disputed, because the sources weigh the size of the added bleeding risk differently.

Marked passages are checked key claims. Unmarked text was not compared individually.

See all differences
Contradiction · critical Is an occasional ibuprofen dose acceptable?
ChatGPT, Claude, Gemini, DeepSeek
No. Avoid NSAIDs entirely unless a doctor explicitly approves.
even single doses can meaningfully increase bleeding risk
Mistral, Grok
A one-off low dose is usually tolerated.
occasional use is generally considered low risk for most patients
Worth verifying: this is a medical decision. Confirm with a doctor or pharmacist before combining the two.

04 · Monitor

Know when the answer changes.

Turn any consensus into a Watch. consens.io reruns the same question on your schedule and alerts you only when the result materially moves.

Consensus Watch Active

Has the EU guidance for general-purpose AI models changed?

WeeklyMaterial changes onlyTelegram
Jul 14
Baseline72/100 model agreement
Jul 21
ChangedNew transparency requirements are now included.
34/100 movement
Consensus changedTelegram · now

The latest run now includes new transparency requirements. The core conclusion moved materially.

Create a Watch

Proof, not promises

We benchmarked the consensus.

On 314 closed-book MMLU-Pro questions, the consensus ranks first overall, and pulls clear exactly where the models disagree.

When the models disagree

Accuracy on the questions where they split · MMLU-Pro

72.1%
70.6%
69.1%
64.7%
58.8%
44.1%
39.7%
consens.io Claude GPT-5.5 Gemini Grok DeepSeek Mistral
#1Highest accuracy overall and on the hard questions. The top of the field is a statistical tie.
+32ptsLead over the weakest model on questions where the answers split.
See the full benchmark

Why consens.io

Built for decisions that need a second view.

Compare

Original answers stay visible

Read each model response side by side before the consensus layer summarizes it.

Verify

Disagreement is marked in place

Where the models pull apart, the sentence itself carries the mark, instead of the doubt being smoothed into one confident answer.

Control

You choose the models

Select providers, switch reasoning modes, and decide which outputs count toward consensus.

Model coverage

One workspace for leading AI providers.

Direct comparison across major model families, so every answer can be checked against independent alternatives.

Get the product view

Compare an answer or keep watching it.

Open app