This policy describes the current Terminal AP release. Updated 8 October 2026.
Data on your device
Watchlists, recent tickers, positions, chart drawings, horizontal levels, research-view preferences, workspace drafts and named layouts, Home layout preferences and last-visit quote comparisons, scoring-timeframe preferences, and inactive alert rules are saved in this browser’s local storage. They are not synchronised to an account or sent to a portfolio service. Search and command histories can be paused or cleared in the search panel; pausing navigation history also clears recently viewed assets. Command history can be disabled independently. Home visit comparisons can be paused or cleared under Home → Customise; pausing navigation history clears these comparisons too. Watchlists and positions are kept. Clearing site data removes these saved items. Trade-planner drafts remain in the current tab.
Saved screeners
Saving a screener requires ChatGPT sign-in. Its name, filters, selected columns and sort settings are stored on the server against your authenticated account ID. Saved screens are private to that account and can be edited, duplicated or deleted in the Screener. Public indexed stock metrics are shared research data and are not linked to your saved screens or account.
Requests and service logs
When you request market research, the ticker, dataset and request parameters are sent to our server. The server requests public market data from providers including Yahoo Finance and, for some earnings fallbacks, Nasdaq. The AP Intelligence Feed also requests public RSS headlines from the Federal Reserve, European Central Bank and Bank of England. It sends only selected ticker symbols and market filters to request news, and never sends position sizes or purchase prices. Refresh comparisons stay in memory during the current visit. When an AI summary connection is enabled, generating a summary sends only public headlines and source metadata to OpenAI, without holdings, purchase prices or personal preferences. AI interpretations are clearly labelled and cached separately from source facts. Home also checks published US release schedules from the Bureau of Labor Statistics, Bureau of Economic Analysis and Federal Reserve. Browser requests also reach the hosting infrastructure, which may process technical information such as IP address, user agent, request path and time for delivery, security and operational logs. Do not treat research queries as confidential.
Public research history
Opening Sentiment can save one server-calculated observation per stock per UTC day. Opening asset research can save a distinct server-calculated AP Score observation when its evidence changes, separated by asset, timeframe and methodology version. Market Pulse can also save public observed market snapshots at most once per 15-minute window. Opening Risk Radar can save one public server-calculated asset or broad-market risk observation per UTC hour, including score drivers, source dates and methodology. Portfolio risk is calculated on this device; position quantities, portfolio weights and portfolio history are not uploaded. Portfolio comparisons cover the current visit. These public research records contain the ticker, score, contributing categories, source dates and methodology. They are shared across visitors and are not linked to an account, visitor identifier, watchlist or position. History records when research was observed; it is not a record of your personal trades.
Opening AP Timeline archives dated public asset events and meaningful changes from the saved AP score and risk observations. Records are shared across visitors and contain no visitor identifiers or personal portfolio data.
Tracking
Anonymous search diagnostics record result counts, no-result and failure outcomes, coarse latency ranges and selected market symbols. They do not add visitor identifiers or tracking cookies, or include raw search text, watchlists or position values. Browser Do Not Track and Global Privacy Control signals are respected. Hosting-provider operational logs may still apply independently. No advertising trackers or cross-site profiles are added. We do not sell position or watchlist data. Saving screens uses ChatGPT sign-in. There is no subscription checkout.
External links
Following a source or company link opens another service with its own privacy practices. Contact information for the site operator has not yet been published. This page will be updated when additional account features, contact channels or analytics are introduced.