In a day, an analyst receives more reviews, issuer reports and prospectuses from Bloomberg Terminal, LSEG Workspace and e-mail than anyone can read. A private model prepares a summary on your topics (risk factors, fees, changes since the previous version) with a reference to the specific page, with no assessments or recommendations.
Built for ISO 27001NIS2 DORA
In an investment management firm, every morning the analyst opens Bloomberg Terminal or LSEG Workspace, where market reviews from several brokers and issuers' quarterly reports are waiting, along with stock exchange announcements about issuer events, and in e-mail there are updated fund prospectuses, key information documents and Morningstar Direct fund reports. The analyst opens a hundred-page prospectus to find out whether the fees and the risk section have changed since the previous version. Some documents stay unread, and a change is noticed only when a client or the compliance officer asks about it. Knowledge of what has changed is held in individual people's heads.
The solution takes documents from the analysts' mailbox, from a shared SharePoint folder that Bloomberg and LSEG material is exported to, and from the Morningstar Direct report export, then passes them to Claude running in a private environment in Amazon Bedrock in an EU region. The model prepares a summary on the topics you set: risk factors, fees and costs, changes to strategy or limits, stock exchange and regulatory announcements, differences from the previous version. Every statement has a reference to a page of the document. The summary is descriptive, with no assessments, forecasts or buy and sell suggestions, because recommendations are given by your licensed people. The analyst reads the summary in a Teams channel, opens only the flagged pages and decides what to do next. We connect to Bloomberg, LSEG and Morningstar through export or an e-mail flow, according to your licence terms. Documents are not used to train the model, and processing is written to a log.
Four steps from input to a record in your system, with human approval wherever there is risk.
The flow watches the analysts' mailbox, a shared SharePoint folder that reviews from Bloomberg Terminal and LSEG Workspace are exported to, and the Morningstar Direct report export. A new document is assigned to an issuer, fund or topic and matched with the previous version, if one is kept in your archive.
A private model reads the document and fills in your summary structure: risk factors, fees, changes to strategy or limits, stock exchange and regulatory announcements, differences from the previous version. Each statement names the page. The model does not assess anything or suggest actions.
The summary appears in the analysts' Teams channel or SharePoint portal next to the original. The analyst opens only the flagged pages and either confirms the summary or flags an inaccuracy. Conclusions and recommendations are prepared by a person; the summary only shortens the path to the source in Bloomberg or LSEG.
Scanned or protected documents whose text cannot be read are flagged and left to the analyst. Every processing step is written to a log with the document version and source. When the topics you track change, the Bloomberg or LSEG export format changes or a new document type appears, we update the summary structure.
Automatic intake of reviews, reports and prospectuses from a mailbox, a SharePoint folder with Bloomberg and LSEG exports and Morningstar Direct reports, with version storage.
A documented set of topics for each document type (broker review, issuer report, prospectus, stock exchange announcement) and a summary format with mandatory source references.
A list of differences between the new version of a prospectus or fund document and the previous version stored in your SharePoint archive.
A processing log with document versions and sources, a quality review of summaries with the analysts and updates to the topic list.
Claude through the API in a private environment in Amazon Bedrock in an EU region + a document flow from Bloomberg Terminal and LSEG Workspace exports, Morningstar Direct reports and e-mail. We connect to these sources through file export or an e-mail flow, according to your licence terms, and we choose the method after a check. Learn more about this service and its limits on the service page.
No. A summary only describes what the document says and what changed, with no assessments or suggestions. Recommendations to clients are prepared by your licensed employees under your firm's procedures.
Yes, but not through direct access to the terminal. We take Bloomberg and LSEG reviews from analyst exports into a SharePoint folder or from e-mail subscriptions, and Morningstar Direct reports from their export, because data redistribution depends on your licence terms. We choose the integration method after a check and document it so that you do not breach your contracts with data providers.
The model runs in a private environment in Amazon Bedrock in an EU region, and documents are not used for training and do not leave the EU. If you need stricter control, we deploy the same solution in the MDP CLOUD GPU cloud in Lithuania (EU).
After a client meeting, the manager writes notes into Salesforce or Dynamics 365 from memory, and some of the agreed actions get lost.
At the end of the month, the portfolio manager prepares dozens of reports and letters from the same Euroclear statements and Excel tables.
A new version of ESMA guidelines, an updated decision of the financial regulator, changed stock exchange rules, a changed internal policy: the…
What is the limit per issuer, who approves a trade above the limit, how do I declare a conflict of interest: the answers are in the policies in…
This service is delivered by specialists with valid international certifications. Their expertise is confirmed by independent organisations, not by us.
AI systems are deployed with protection against data leakage and model abuse.
The AI security programme and its controls are managed following the ISACA methodology.
AI risks are identified and managed like any other IT risk.
We reply within 1 business day
