Overview
Reduce compliance risk. Speed up investigations.
Agentic result investigation is an optional feature that helps you investigate disclosure results faster and with greater confidence than ever before.
It brings together relevant historical context, filing activity, position changes, Denominator Discovery and regulatory information, so you can quickly understand why a result was triggered.
Please note that Denominator Discovery is currently being tested by early adopters, with wider availability to be announced.
With agentic result investigation, you can:
Reduce the risk of missed or incorrect disclosures by consistently surfacing key information.
Save time and effort without sacrificing accuracy.
Free up your team’s time to focus on the decisions, not the data gathering.
Agentic result investigation supports your investigation while you stay in full control. It doesn’t make decisions or take action on your behalf.
The key outcome is a reduction in disclosure investigation effort by around 25% (as seen in an early testing cohort), allowing you to focus more time on analysis and decision-making.
AI Capabilities
Result Investigation | Available Now
Reduce the risk of missed or incorrect disclosures by surfacing key information needed to validate a result.
Investigating a disclosure result often requires checking multiple sources of information, increasing the risk that important details are overlooked.
Result Investigation analyses the information needed to validate a result, including:
Why the disclosure was triggered.
Relevant filing and disclosure history.
What has changed since the previous result.
Which positions contributed to the change.
Key information needed to validate the result.
By bringing this information together in one place, you can spend less time gathering evidence and more time assessing the outcome.
The key benefit is more accurate, consistent investigations completed in less time.
Denominator Discovery | Coming Soon
Reduce the risk of missed or incorrect disclosures by ensuring the correct denominator is used.
Validating denominator data often requires manually searching information across multiple public sources.
Denominator Discovery gathers denominator information from issuer websites, regulatory filings, and exchange publications and compares it with the data used in our Monitoring service.
When available, you will be able to:
Review denominator information from multiple public sources, with links to each source.
Automatically calculate the variance between the data sources and the one used in the rule calculation.
Spot discrepancies that may need further review.
Spend less time on manual validation.
Instead of manually collecting and comparing denominator data across multiple sources, you can quickly identify and investigate potential issues before a disclosure is made.
The key benefit is faster denominator validation and greater confidence that the correct denominator is being used before a disclosure is made.
How It Works
Open a Shareholding Disclosure result in your environment.
Navigate to the AI Insights box located to the right of the Assets Table.
Review the AI-generated summary and contextual information.
Validate the findings and supporting details.
Complete your investigation and determine the appropriate next steps.
Frequently Asked Questions (FAQ)
Q: Can I choose not to use Results Insights?
A: Yes. Results Insights is an opt-in feature; you can choose whether to use it. Simply head over to: Admin and settings > Setup > Agreements in your environment. Additionally, approval can be withdrawn at any point in time.
Q: What model is being used?
A: The Result Insights feature uses Anthropic’s Claude Haiku 4.5 and Sonnet 5.0.
Q: Can I select the model I want to use?
A: No, given the complexity and the guardrails that need to be put in place, you have to rely on the selected model.
Q: What guardrails do you have in place?
A: The models we use are hosted on AWS Bedrock. AWS provides native industry-leading guardrail solutions. Specifically, we use:
Content filters (help detect and filter harmful user inputs and model-generated outputs).
Word filters (namely, the profanity filter to block profane words).
We mask any personal information (names, email addresses).
Q: Is client data used to train the model?
A: No, client data is not used to train the model powering the Results Insights.
Q: Is my data secure?
A: AI Insights operates within a controlled environment and is protected by standard platform security controls and Amazon Bedrock guardrails.
Q: What denominator types are supported?
A: Denominator Discovery currently supports Total Shares Outstanding (TSO) and Total Voting Rights (TVR).
Q: What denominator sources are retrieved?
A: Denominator Discovery retrieves denominator information from publicly available Issuer, Regulatory, Exchange, and Other sources. "Other" sources may include financial news articles. The source type is displayed alongside each retrieved denominator value.
Q: How does Denominator Discovery prioritise sources?
A: Denominator Discovery prioritises sources in the following order: Issuer > Regulatory > Exchange > Other. If no denominator information can be found for a source, no data will be returned in the UI.
Q: Does Denominator Discovery guarantee the denominator is correct?
A: Denominator Discovery assists with research by identifying publicly available denominator information, but users remain responsible for reviewing and validating the denominator before using it in a disclosure filing.
Q: What date does the denominator relate to?
A: Denominator Discovery returns the denominator value at the time of the search, together with the source information and publication date. This allows users to know when the denominator was published.
Q: Can I view the source used to identify the denominator?
A: Yes. A URL link is displayed for each source, allowing users to review the supporting evidence.
Q: Why can't I find a denominator for a particular issuer?
A: Denominator Discovery relies on real-time retrieval of publicly available information; occasional failures can occur due to temporary website outages or access restrictions.
Q: What source information is displayed?
A: For a given denominator, Denominator Discovery displays the extracted value, source name, source type, and supporting source link where available.
Learn More
For detailed information on our approach to AI, including governance and controls, please visit our AI Policy Hub.
