Question 1

How does Multi-Model Consensus work in Xperbot?

Accepted Answer

Xperbot's Multi-Model Consensus Engine runs a 4-stage cross-model peer review pipeline. Stage 1: Google Gemini 3.8 Flash scours live Google Search indices to construct an evidence-grounded factual baseline. Stage 2: OpenAI GPT-4o audits the synthesis for deductive consistency and statistical validity. Stage 3: Anthropic Claude 3.5 Sonnet scrutinizes epistemic nuance, unstated assumptions, and cognitive bias. Stage 4: Xperbot computes a unified Triangulation Index with verified consensus pillars and divergence points.

Question 2

Which AI models are involved in Xperbot's consensus verification?

Accepted Answer

Xperbot triangulates consensus across three leading frontier AI models: Google Gemini 3.8 Flash (live search grounding engine), OpenAI GPT-4o (deductive reasoning and numerical consistency validator), and Anthropic Claude 3.5 Sonnet (epistemic nuance and cognitive bias auditor).

Question 3

How does the Consensus Radar work in Xperbot?

Accepted Answer

The Consensus Radar analyzes all grounded web sources and frontier AI models retrieved for your query, computes an overall source agreement score (0-100%), and maps out explicit points of contention and divergence between different publications.

Question 4

What is the Scenario Sandbox in Xperbot?

Accepted Answer

The Scenario Sandbox transforms factual claims and forecasts into interactive parameter models where users can adjust grounded variables (e.g. growth rates, interest rates, timelines) with dynamic sensitivity dials to see real-time recalculations.

Question 5

What does the Volatility Meter do?

Accepted Answer

The Volatility Meter evaluates the expected half-life and decay rate of factual claims, categorizing them as permanent, stable, high-volatility, or transient, with live one-click temporal verification checks.

Question 6

What is the Adversarial Red Team Auditor?

Accepted Answer

The Red Team Auditor stress-tests synthesized findings by proactively executing adversarial searches to locate counter-evidence, flawed premises, and dissenting peer viewpoints.

Question 7

How does Xperbot eliminate AI hallucinations?

Accepted Answer

Xperbot queries Google Search in real time for every inquiry, cross-verifying web sources before generating responses. Every assertion is backed by verifiable web citation tables with live URLs.

Question 8

Can I export my research conversation to PDF or Markdown?

Accepted Answer

Yes. Xperbot includes a built-in Export tool in the chat header that allows one-click downloads of your entire research session as a formatted multi-page PDF report or structured Markdown (.md) document.