Overview | Telephos Trust Center

Telephos

Telephos is an AI-powered platform that processes call recordings, transcripts, and team communications like Slack messages to deliver actionable business insights and optimize workflows. Security is core to everything we build. Our SOC 2 audit is complete, covering Security, Availability, and Confidentiality with encryption at rest and in transit, role-based access control, continuous vulnerability monitoring, and rigorous vendor management.

Have a security question or need to report a concern? Contact us at security@telephos.ai. We respond within one business day.

Controls

View all

Infrastructure security

Unique production database authentication enforced
Encryption key access restricted
Unique account authentication enforced

View 13 more controls

Organizational security

Production inventory maintained
Confidentiality Agreement acknowledged by employees
Performance evaluations conducted

View 1 more control

Product security

Control self-assessments conducted
Data transmission encrypted
Vulnerability and system monitoring procedures established

View 1 more control

Internal security procedures

Configuration management system established
SOC 2 - System Description
Board oversight briefings conducted

View 18 more controls

Data and privacy

Customer data deleted upon leaving

Data collected

Customer personally identifiable informationCollected

Employee personally identifiable informationCollected

Credit card informationNot collected

Personal health informationNot collected

Subprocessors

View all

Anthropic

• Large Language Model Service for Data Analysis

Telephos uses Anthropic's API (Claude) for portions of analysis (call transcript reasoning and complex business-insight extraction). Per Anthropic's standard API terms, content submitted via the API is not used to train Anthropic's models. Excerpts of transcript text and derived prompts are sent; no PII beyond what appears in call content. Processed in US regions. Learn more at anthropic.com.

USA

OpenAI

• Large Language Model Service for Data Labeling

Telephos uses OpenAI's API for portions of analysis (transcript classification and summarization). Per OpenAI's standard API terms, content submitted via the API is not used to train OpenAI's models. Excerpts of transcript text and derived prompts are sent; no PII beyond what appears in call content. Processed in US regions. Learn more at openai.com.

USA

Render

• Application Hosting and Deployment

Managed application hosting and PostgreSQL database infrastructure. Processes call transcript text, customer account metadata, user identifiers, and integration tokens. Data resides in Render-managed infrastructure in the United States. Selected for SOC 2 Type II audited controls, managed Postgres with at-rest encryption, and private networking between services. Learn more at render.com.

USA

Vercel

• Frontend Application Hosting

Hosts Telephos's frontend web applications (marketing site, customer web app). Processes in-flight web requests including authenticated session data. No persistent storage of customer data on Vercel; persistent data resides in Render-managed infrastructure. Processed in US regions.

USA

FAQ

View all

Do you encrypt data at rest?

Yes. All customer data is encrypted at rest using AES-256 encryption via our managed PostgreSQL database. Data is also encrypted in transit using TLS 1.2+ for all API and web traffic. Sensitive fields like API credentials are additionally encrypted at the application layer using AES-256-GCM before storage.

Where are your servers located?

Telephos infrastructure is hosted on Render, a SOC 2-certified cloud platform built on AWS and GCP. All production services and databases run in US-based data centers. We have no on-premises servers. We are fully cloud-native.

Telephos