Research Signal Disclosure

Purpose of This Disclosure

This page explains how Kerdos AI presents research signals and forecast scenarios, and what contextual information should accompany generated outputs where available.

Some Kerdos AI outputs may discuss specific financial instruments, sectors, markets, rankings, scenarios, or model classifications. Depending on their content, presentation, distribution, and applicable law, such outputs may be interpreted as information suggesting an investment strategy or as investment-recommendation-like content.

Kerdos AI therefore aims to present research outputs with clear warnings, timestamps, source context, methodology notes, assumptions, rating explanations, and conflict-of-interest disclosures where practical.

Signal Labels

Signal labels are research classifications generated by Kerdos AI’s models and methodology. They are not buy, sell, hold, or trade instructions.

Signal labels may include:

• Very Positive Signal;
• Positive Signal;
• Neutral Signal;
• Negative Signal;
• Very Negative Signal.

A positive signal does not mean that an instrument is suitable for any user or that it will increase in value. A negative signal does not mean that an instrument must be sold, avoided, or will decrease in value.

Signal labels should be reviewed together with the relevant data context, assumptions, methodology notes, risk factors, forecast horizon, confidence indicators, and the user’s own independent research.

Forecast Scenarios and Scenario Price Targets

Kerdos AI may display forecast scenarios, upside scenarios, downside scenarios, base scenarios, or scenario price targets.

These are hypothetical model-generated outputs based on available data and assumptions. They are not guarantees, promises, or reliable predictions of future prices.

Scenario price targets are not instructions to buy, sell, hold, or trade. They should be understood as research scenarios, not expected outcomes.

Universe Rankings

Kerdos AI may rank instruments within a selected research universe. A ranking is a relative model output within a defined universe, selected methodology, and point in time.

A higher-ranked instrument is not necessarily suitable for any user and is not guaranteed to perform better than a lower-ranked instrument.

Rankings can change when the universe, filters, data, assumptions, model version, methodology, or market conditions change.

Expected Output Metadata

Where available and applicable, generated outputs should display or make available the following information:

• producer identity, such as Kerdos AI;
• generated-at timestamp;
• data-as-of timestamp;
• price-as-of timestamp, where price data is shown;
• relevant market, exchange, sector, or universe;
• forecast horizon or research horizon;
• signal label and explanation of its meaning;
• confidence indicator and explanation of its meaning;
• source categories used;
• material source references where available;
• distinction between factual information, model output, and AI-generated interpretation where practical;
• key assumptions;
• methodology summary;
• known limitations;
• main risk factors;
• conflict-of-interest disclosure;
• AI-generated content notice where applicable;
• previous output or change notice where applicable.

Unavailable Metadata

If backend data or source context is unavailable, Kerdos AI should display clear unavailable-state copy rather than fake values.

Examples include:

• “Data-as-of timestamp unavailable.”
• “Source category unavailable.”
• “Methodology summary unavailable.”
• “Conflict status not checked.”
• “Price timestamp unavailable.”

The absence of metadata should be treated as a limitation of the output.

Risk Warning

Research signals, rankings, forecast scenarios, confidence indicators, price-target scenarios, and AI-generated summaries may be wrong, incomplete, delayed, or based on inaccurate data.

Users should not rely on Kerdos AI as the sole basis for investment decisions. Users remain responsible for their own research, decisions, and investment outcomes.

Conflicts of Interest

Where a material conflict exists or is reasonably expected, Kerdos AI aims to disclose it clearly and close to the relevant content where practical.

Potential conflicts may include, for example:

• positions held by Kerdos AI, its owner(s), employees, contractors, or affiliates;
• issuer compensation;
• sponsored content;
• broker relationships;
• affiliate or referral compensation;
• data-provider relationships;
• commercial relationships that may affect, or appear to affect, research outputs.

For more information, please review the Conflicts of Interest disclosure.

Updates, Changes, and Withdrawal

Kerdos AI content may be updated, replaced, corrected, withdrawn, or become outdated.

Outputs generated at different times may not be directly comparable because data, assumptions, models, methodology, market conditions, and source coverage may change.

No Personalised Advice

Kerdos AI does not assess any user’s full financial situation, investment objectives, investment experience, risk tolerance, loss capacity, tax position, liquidity needs, or legal restrictions.

Filters, watchlists, selected horizons, markets, sectors, or other settings are used to organise research outputs only. They do not create personalised investment advice.

General Nature of This Disclosure

This disclosure is general in nature and may be updated from time to time as Kerdos AI’s features, models, methodology, legal structure, business model, or applicable requirements evolve.

Last updated: 2026-06-07