Does it block the trunk when it detects fraud?

Not without approval. Blocking drops legitimate customers' calls if the indicator is wrong, so the agent documents the pattern with the CDR behind it and the blocking decision stays with the team — with the evidence already prepared to act fast.

What kinds of fraud can it detect?

Patterns that show up in the detail records: spikes outside the historical profile, unusual destinations, high volumes of short calls, concentration in high-tariff ranges. The agent shows the pattern and the period, and whoever knows the operation confirms whether it is fraud or a legitimate campaign.

Does it execute the portability?

No. It tracks the request, warns before the regulated deadline lapses, and organizes the paperwork. Execution follows the operator's procedure with the entities involved.

Does it submit filings to the regulator?

No. It keeps the calendar, organizes the inputs, and flags what is missing before the deadline. Submission and responsibility for the filing stay with the operator.

How does wholesale reconciliation work?

It cross-checks measured traffic against what the partner billed and isolates the gap by route and period, with the reference CDR. That turns a dispute based on impressions into a dispute with numbers.

Is call data used to train models?

No. Content processed by Secure AI Agents is not used to train UPX models or third-party models.

Can we audit what the agent did?

Yes. Every action is logged: what was queried, what was proposed, when, in which system, and under which permission. The history stays available for review and auditing.

Secure AI Agent · Telecom

AI applied to
the traffic you bill

A specialist configured to detect fraud patterns in CDRs, keep portability deadlines on the calendar, reconcile wholesale traffic, and organize regulatory filings.

Get started See pricing

+1,400 integrations available

Companies that trust UPX

Capabilities

What the Telecom Agent
can do

Main areas of work for the Telecom and Voice Operator AI Agent in your operation.

Fraud detection

Finds the suspicious pattern in the CDR and shows the evidence.

Portability deadlines

Tracks the request and warns ahead of the regulated deadline.

Wholesale reconciliation

Cross-checks billed against measured traffic and isolates the gap.

Regulatory calendar

Monitors obligations and organizes the filing inputs.

Call quality

Correlates complaints with routes and flags the bad segment.

Invoice disputes

Pulls the period's CDR and builds the case for the figure.

Skills

Capabilities that compose the specialist

Skills add specific capabilities to the Secure AI Agent according to the processes it needs to execute.

Explore skills

CDR analysis: Sweeps the detail records and isolates the outlier pattern.
Traffic reconciliation: Compares measured against billed and flags the deviation.
Portability calendar: Monitors the request and warns before the deadline lapses.

How it works

From the raw CDR
to contained fraud

Connect your sources, define what the agent can do, and operate with evidence before the loss closes the month.

Connect the traffic and the controls: The agent works with what the platform already records.Connect the CDRs, the customer base, the portability controls, and the wholesale billing so the Secure AI Agent has access only to what it needs.
Ask for the work: Talk to the agent in natural language.Ask for a fraud sweep over the period, the portability requests about to lapse, or the wholesale reconciliation through the available channels. The agent applies the configured skills and cites the source CDR.
The agent detects. You decide.: From detection to blocking, with control.The agent analyzes, reconciles, and documents within the defined permissions. Blocking trunks, cutting routes, and submitting regulatory filings stay with the responsible team.

Integrations

Connected to the platform and the controls

The Telecom Agent can query your traffic and run tasks in the systems your team already uses.

Twilio

Explore integrations

Flow

What goes in,
what the agent does, and what comes out

From raw detail records to a defensible operation, following the operator's procedures and permissions.

Inputs

Call detail recordsCDR
Wholesale billingCSV
Portability requestsLOG

Processing

Secure AI Agent

Processing the task

Read
Correlate
Isolaterunning
Document

CDR citedProcedure verified

Output

Completed

Analysis prepared

Fraud pattern with the evidence
Wholesale gap isolated
Portability about to lapse flagged

Control

You define how far the Agent can act

Different actions can operate with different autonomy levels, always within the operator's procedures.

1
Query
Reads CDRs and controls and answers with the evidence.
2
Prepare
Isolates the suspicious pattern and builds the reconciliation.
3
Request review
Waits for approval before acting on the traffic.
4
Execute
Performs the action within the defined limits.

Levels are configured per type of action. Blocking trunks, cutting routes, executing portability, and submitting regulatory filings always stay with the responsible team, because they affect customers and legal obligations.

Explore governance

Get started

Put a Secure AI Agent to work.

Start on the platform or choose the plan that fits the pace of your operation.

Get started See plans

Security and compliance

Security that can be verified.

Certifications and attestations

UPX maintains SOC 2 Type II and ISO 27001, with independent audit over its information security controls.

Explore security

Privacy and regulation

LGPD: Operations follow Brazil's Law 13.709/2018. In AI Agent contracts, UPX acts as data processor; the legal basis remains with your company.
Zero Data Retention: A product policy, not a certification: with compatible providers and configurations, processed content is not retained after execution.
Retention and deletion: The retention policy is defined by contract. Once the contract ends, data is deleted within the agreed period.

Frequently asked questions

Common questions about the Telecom Agent

What operators usually ask before putting an agent on the traffic.

Does it block the trunk when it detects fraud?
Not without approval. Blocking drops legitimate customers' calls if the indicator is wrong, so the agent documents the pattern with the CDR behind it and the blocking decision stays with the team — with the evidence already prepared to act fast.
What kinds of fraud can it detect?
Patterns that show up in the detail records: spikes outside the historical profile, unusual destinations, high volumes of short calls, concentration in high-tariff ranges. The agent shows the pattern and the period, and whoever knows the operation confirms whether it is fraud or a legitimate campaign.
Does it execute the portability?
No. It tracks the request, warns before the regulated deadline lapses, and organizes the paperwork. Execution follows the operator's procedure with the entities involved.
Does it submit filings to the regulator?
No. It keeps the calendar, organizes the inputs, and flags what is missing before the deadline. Submission and responsibility for the filing stay with the operator.
How does wholesale reconciliation work?
It cross-checks measured traffic against what the partner billed and isolates the gap by route and period, with the reference CDR. That turns a dispute based on impressions into a dispute with numbers.
Is call data used to train models?
No. Content processed by Secure AI Agents is not used to train UPX models or third-party models.
Can we audit what the agent did?
Yes. Every action is logged: what was queried, what was proposed, when, in which system, and under which permission. The history stays available for review and auditing.

Secure AI Agent

Bring a Secure AI Agent to your carrier operation

Talk to our specialists and see how to adapt this AI Agent to your processes, systems, and needs.

Talk to specialists

AI applied to the traffic you bill

What the Telecom Agent can do