PricingGetting Started
Home/Answers/ai-agent-rumor-vs-fact
Context & Meaning·0.07ms Verification

Why AI Agents Act on Guesses and Rumors Like Proven Facts

Why language models confuse "might happen" with "did happen" and act on unverified rumors.

Direct 40-Word Answer

AI agents confuse guesses with facts because AI models represent all assertions as flat text. An informal comment saying "I think John might resign" has the same weight as an official resignation letter. Exogram verifies sources, blocking actions based on unverified rumors.

What Actually Happened in the Real World

An HR automation bot read an internal discussion channel where a manager brainstormed restructuring scenarios. The bot concluded that five employees had been terminated, automatically revoked their Okta access, and triggered severance payout workflows.

Why Polite Prompts Like “Please Don't Do This” Fail

System prompts telling an AI to "distinguish between facts and opinions" fail because models cannot verify external real-world ground truth without cryptographic evidence.

The Sub-Millisecond Code Fix

Exogram attaches evidentiary verification levels to every piece of context. Tool calls requiring high-privilege actions (like revoking account access) can only be triggered by facts tagged with verified official authority.

TYPESCRIPTExecution Boundary Gate (0.07ms)
import { ExogramAdmissibility } from '@exogram/sdk';
const adm = new ExogramAdmissibility();

export async function processTermination(employeeId: string, context: { text: string; source: string; verified: boolean }) {
    // Only verified HR system records are admissible for termination actions
    adm.requireAdmissibility(context, { minEvidenceLevel: 'OFFICIAL_SIGNED_RECORD' });
    return hrSystem.terminate(employeeId);
}

Frequently Asked Questions

Why do AI models confuse guesses with confirmed facts?

Language models treat all text as flat words. They cannot tell the difference between someone saying "I think John might resign" and an official signed resignation letter unless source verification is enforced before tools run.

Stop AI Mistakes Before They Execute

Exogram sits directly between your AI model and your tools. Set up in 10 seconds inside Claude, Cursor, ChatGPT, or your own code.