The application
AI Copilot
Ask about your portfolio, understand an alert and control AI usage.
On this page
Enable AI in Settings, then ask the portfolio chat a question or open a page’s copilot. For example: “Why has my allocation changed?” or “What does this alert mean?”
The daily briefing and scheduled reports can also generate analyses without a separate request. Disabling AI stops these generations and on-demand features.
Fifteen modes
Three features on every plan, eleven more on Pro and one on Premium. Premium is unavailable for purchase; access depends on the plan assigned to your account.
Control your allowance
Your plan sets usage limits. You can choose a lower personal limit and change it later within your plan’s allowance.
Your data as context
When AI is enabled, data relevant to the analysis is used to explain your situation.
One switch
Turning AI off in Settings stops every mode, including the ones that run on their own.
Daily briefing
- What it does
- An executive summary of your portfolio with priorities, a watchlist and notable changes, generated once a day and on demand from the dashboard.
- Plan
- Free
Portfolio chat
- What it does
- Questions about your own data and about how Patrima works.
- Plan
- Free
Explain this alert
- What it does
- Why an alert fired, with the numbers, two or three options including doing nothing, and what to watch.
- Plan
- Free
Allocation copilot
- What it does
- Reads your drift against targets and comments the rebalancing options.
- Plan
- Pro
Accounts copilot
- What it does
- Reads the structure of your accounts and their concentration.
- Plan
- Pro
Strategy copilot
- What it does
- Comments a trading plan and its signals.
- Plan
- Pro
Backtest copilot
- What it does
- Reads a backtest and its limits.
- Plan
- Pro
Projection copilot
- What it does
- Reads a projection and the sustainability of withdrawals.
- Plan
- Pro
Stable Yield copilot
- What it does
- Explains an optimizer proposal against your current positions and their scores.
- Plan
- Pro
Pool copilot
- What it does
- Reads one pool: its rate, its score, its risks.
- Plan
- Pro
News copilot
- What it does
- Reads the news stories against your holdings.
- Plan
- Pro
Market brief
- What it does
- The state of the market, independent of your holdings.
- Plan
- Pro
Asset recap
- What it does
- The last 7 days of news on one asset.
- Plan
- Pro
Portfolio report
- What it does
- The narrative of a report, written around figures computed without it.
- Plan
- Pro
Advanced strategy copilot
- What it does
- Reads a leveraged strategy and its stress scenarios.
- Plan
- Premium
AI settings show your limits and estimated monthly usage. The budget shown in dollars measures your included usage allowance; it is not an additional invoice. Your effective limit is the lower of your personal setting and your plan’s allowance. You can change your setting within that allowance.
AI features pause when a limit is reached. Check the message shown: an hourly limit resets sooner than a monthly budget. Chat messages can contain up to 500 characters.
AI must be enabled to analyse your portfolio. Relevant information provides context for each feature: positions, allocation, risks or news, depending on the request. The daily briefing and scheduled reports also use this context.
In Settings, Analysis mode produces explanations about your data. Demonstration mode displays predefined examples; it does not analyse your portfolio. Standard and In-depth settings let you choose the analysis level. See the privacy policy for data processing terms.
The portfolio chat answers questions about your portfolio and how to use Patrima. Help answers include links to this documentation. Check the figures and linked explanations: AI answers can contain errors.
New topic starts a fresh conversation. With Keep chat history enabled, the conversation can be recovered for two hours after a reload. Without it, the conversation is limited to the open session. Interrupted responses are marked on screen.
Explain this alert helps you understand what triggered it and what to monitor. Helpful and Not helpful let you give feedback on a response. An analysis marked Not helpful is hidden and can be regenerated on the next request.
Check before deciding
The copilot explains figures and presents possibilities. Its answers can be incomplete or incorrect. Check the assumptions and your circumstances; you carry out any operations yourself on your platforms.
Frequently asked questions
How do I enable the AI features?
Open AI settings, enable AI and select Analysis. Demonstration lets you explore the screens with predefined examples. Usage limits and chat history are configured in the same place.
What can I ask the portfolio chat?
Ask for explanations of your positions, allocation, yields or alerts, or for help using Patrima. Use New topic to start a different conversation.
Why do I see "Monthly budget reached" or "Rate limited, try again later"?
You have reached your monthly allowance or hourly limit. Check the effective limit in AI settings. If you chose a lower personal limit, you can raise it within your plan’s allowance. Otherwise, wait for the limit to reset.
What data is sent to the AI, and when?
Only when AI is enabled, and only the portfolio context relevant to each request. Two cases send data: when you use an AI feature (open a copilot, send a chat message, generate a report), and the daily briefing and scheduled reports, which run on their own once AI is on. Disabling AI stops all of it.
What is Demonstration mode for?
It shows predefined responses so you can try the AI screens. These examples do not analyse your portfolio and do not use your AI budget.
What does "Explain this alert" do?
On any notification, it asks the copilot to put the alert in context: why it fired (with the numbers involved), 2 or 3 numbered options where doing nothing is always one of them, and what to watch next. It uses your portfolio data, the related position and recent news. The explanation is informational, never an instruction.
What does the Helpful / Not helpful feedback do?
Feedback helps assess responses. An analysis marked Not helpful is hidden; your next request can generate a new one.
See also