# PolyAI — AI Agents & Assistants

PolyAI by PolyAI Limited holds a panel rating of 2/5 from 6 AI judges (category median 3/5), as of 2026-10-28. Provenance: UK / wider Europe; sovereignty: 1 of 4 dimensions proven.

- Vendor: PolyAI Limited (https://poly.ai)
- Category: [AI Agents & Assistants](https://whats-best.ai/software/ai-agents.md)
- Provenance (editorial classification): UK / wider Europe
- Canonical page: https://whats-best.ai/software/ai-agents/polyai
- Alternatives: https://whats-best.ai/software/ai-agents/polyai/alternatives ([markdown](https://whats-best.ai/software/ai-agents/polyai/alternatives.md))
- Data as of: 2026-10-29 · evaluated 2026-10-28 · oldest source captured 2026-10-26

**AI-generated.** Every score, rationale and verdict here is written by a panel of AI judges from public evidence — an analysis, not a hands-on test. Every fact quotes the vendor page it was read from. No score can be bought. How scores are made: https://whats-best.ai/methodology

## At a glance

- Vendor: PolyAI Limited
- Headquarters: Not stated on the vendor’s pages as of 26 Oct 2026
- Data hosting: Not stated on the vendor’s pages as of 26 Oct 2026
- Free plan: Not stated on the vendor’s pages as of 26 Oct 2026

## The panel's verdict

PolyAI is an enterprise voice AI agent platform for customer service, powered by the vendor's proprietary Raven model, and the bench's scores are tight: no criterion spans more than one point. It is strongest on integrations and knowledge and agent capabilities, with 130+ integrations across telephony, CRM and vertical systems, named partners including Salesforce, Zendesk and Twilio, OAuth 2.0, a self-serve Connect Portal, an Agent Developer Kit, and test calls, simulated conversations and A/B tests on live traffic at no extra charge. It is weakest on pricing transparency, where the pricing page publishes no figures and routes buyers to "Talk to our team", and on admin and access, flat at 2, where we found no public information on single sign-on, roles or audit logs. Data use and training and sovereignty stay low: we found no public information on whether conversations train models, even as Raven is advertised as "trained on 1B+ enterprise conversations", and hosting is stated only as AWS, with no server region or published subprocessor list. The judges show no genuine split.

**Speaks for it:**
- 130+ pre-built integrations across telephony, CRM, productivity and vertical systems, with named partners including Salesforce, Zendesk, NICE, Five9, Twilio and Verizon Business.
- Managed OAuth 2.0 with automatic token refresh, custom objects and field mappings, a self-serve whitelabelled Connect Portal, open APIs and an Agent Developer Kit (CLI tool and Python package).
- Test calls, simulated conversations and A/B tests on live traffic are included at no extra charge, with real-time latency and cost monitoring per agent.
- Use is priced per minute and billed per second of connected time, with no minimums and no rounding up to the minute, and new Dialog-RSN-1 API accounts get $100 in free credit for a limited time.
- SOC 2 and ISO 27001 certifications are published, with regular audits and third-party penetration testing.

**Held against it:**
- We found no public information on whether conversations, prompts or outputs are used for training, while Raven is advertised as 'trained on 1B+ enterprise conversations'.
- The pricing page publishes no figures — 'Flexible pricing … Talk to our team' — and we found no public information on per-minute rates, tiers, usage limits or minimum commitments.
- We found no public information on single sign-on, SCIM provisioning, roles, group-based permissions or audit logs.
- Hosting is stated only as AWS with no server region given, we found no published subprocessor list, and the privacy policy mentions service providers with operations 'including in the United States'.
- Retention is publicly stated only for client administrator records — kept for six years after the agreement — with no public information on retention of conversations or call logs.

**Best for:**
- You run an enterprise contact centre on telephony and CRM stacks such as Salesforce, Zendesk, NICE, Five9 or Twilio and want voice agents that act inside those systems.
- You want to test before scaling — test calls, simulated conversations and A/B tests on live traffic are included at no extra charge.
- You route calls over SIP or PSTN and want context handed back to human agents via SIP headers.
- Your team wants to build agents without code from a plain-language description, or programmatically via the CLI tool and Python package.

**Avoid if:**
- Your finance team needs to forecast an annual cost from public pages — the pricing page publishes no figures and routes buyers to 'Talk to our team'.
- You need written answers, before signing, on whether customer conversations train models and how long transcripts are kept.
- You need contractual accuracy guarantees — the terms provide AI Outputs 'as is' and state they 'may contain inaccuracies'.
- Your rollout cannot proceed without documented single sign-on, provisioning, roles and audit logs.

## Scores by judge and criterion

Scores out of 5, each judge scoring independently against the published rubric (v1). "n/r" = not rated: the vendor publishes no price, or the judge found no public information on that criterion.

| Criterion | The Head of IT | The Data Protection Officer | The Automation Lead | The Customer-Service Director | The Works Council Member | The Skeptic |
| --- | --- | --- | --- | --- | --- | --- |
| Data use, training & retention | 1.5/5 | 1/5 | 1/5 | 1.5/5 | 1/5 | 1/5 |
| Models & processing location | 2/5 | 2/5 | 2/5 | 2/5 | 2/5 | 1.5/5 |
| Admin, identity & access controls | 1/5 | 1/5 | 1/5 | 1/5 | 1/5 | 1/5 |
| Integrations, connectors & knowledge | 3.5/5 | 3.5/5 | 3.5/5 | 4/5 | 3.5/5 | 3.5/5 |
| Agent capabilities & guardrails | 3.5/5 | 3/5 | 3/5 | 3.5/5 | 3/5 | 3/5 |
| European sovereignty | 2/5 | 1.5/5 | 1.5/5 | 1.5/5 | 1.5/5 | 1.5/5 |
| Pricing transparency | 1/5 | 1/5 | 0.5/5 | 1/5 | 1/5 | 1/5 |
| **Weighted verdict** | **2/5** | **1.5/5** | **2/5** | **2.5/5** | **1.5/5** | **2/5** |

### The judges

- **The Head of IT** (v1.0): Has to roll an AI assistant out to eight hundred staff without a second shadow-IT wave. Wants SSO and SCIM, groups that decide who gets which model, agent and connector, answers that respect SharePoint permissions, audit logs for the SIEM, and a deployment option that is not "trust us".
- **The Data Protection Officer** (v1.0): Signs the DPA and gets asked where a prompt goes. Wants every model provider named as a subprocessor with its region, training excluded for all of them, retention the company controls, and an EU-only switch that also covers agents and connectors — not just the chat window.
- **The Automation Lead** (v1.0): Builds the agents that file tickets, draft offers and chase approvals across the ERP, CRM and wiki. Wants tools and actions with scoped permissions, MCP and an API, approval steps before anything is sent, run logs, and a way to test an agent before the whole company uses it.
- **The Customer-Service Director** (v1.0): Runs a contact centre with forty agents and wants half the routine calls and chats answered by an AI agent. Wants handover to a human with the full context, actions in the order and ticket systems, evaluation before go-live, and a price per conversation the CFO can forecast.
- **The Works Council Member** (v1.0): Has a co-determination right the moment a tool can monitor staff, and an AI assistant logs every prompt. Wants usage reports that can be aggregated or pseudonymised, prompt logging rules set out in writing, a human approving what agents do to colleagues, and no training on staff input.
- **The Skeptic** (v1.0): Has read "European AI" on a wrapper that sends every prompt to a US model, "agents" that are saved prompts, and "we never train on your data" with no word on the model provider behind it. Asks which model answers, where it runs, what it may do unsupervised, and what a thousand seats cost.

## European sovereignty — proven facts

Built only from facts on the vendor's own pages; a dimension we could not prove is left open, never scored as zero. Headline: 1 of 4 dimensions proven.

| Dimension | Finding | Points | Source |
|---|---|---|---|
| Legal entity | Not determined | — | — |
| Ownership | Not determined | — | — |
| Data residency | Not determined | — | — |
| Subprocessors | US CLOUD Act reach | 0/2 | [S3] |

## Key facts

Each fact is the vendor's own statement, cited to the source page it was read from ([S*n*] → Sources).

**Compliance**
- Certifications: SOC 2 · HIPAA · GDPR · PCI DSS [S2]
- Certifications: SOC 2 · ISO 27001 [S4]
- Children age: 13 [S6]
- Complaint UK ico: Information Commissioner's Office [S6]
- Cookie consent: yes [S6]
- ISO 27001: yes [S3]
- Ncsc aligned: yes [S3]

**Data**
- Opt out email: no · yes [S6]
- Restricted submissions: No confidential, sensitive or regulated data (e.g. PHI, payment card data) via demo forms [S5]
- Retention administrators: client administrators · 6 [S6]

**Hosting**
- Aws: AWS · yes [S3]

**Integrations**
- API: yes · yes [S9]
- API optional initial launch: API integrations are needed only to connect the voice assistant to back-end systems; many calls can be automated without them [S7]
- Built in monitoring: yes · yes · yes [S9]
- Catalog categories: Telephony · CRM and knowledge · Productivity and collaboration · Vertical systems [S9]
- Connect portal: self-serve Connect Portal, whitelabelled, inside Agent Studio · yes [S9]
- Count: 130 [S9]
- Custom field mapping: yes · yes · yes [S9]
- Named partners: Amazon Web Services · Five9 · NICE · Salesforce · Twilio · Verizon Business · Zendesk [S2]
- Named tools: Amazon Web Services · Five9 · NICE · Salesforce · Twilio · Verizon Business · Zendesk [S3]
- Oauth: yes · yes [S9]
- Surface: pre-built connectors, API and MCP server support [S4]
- Surface: Integrations, Technology, Data and Insights, Security and Developers sections listed for the platform [S8]
- Telephony sip pstn: SIP or PSTN connection is all that's required to route calls between the voice assistant and the contact center team [S7]
- Zero external dependencies: no [S9]

**Legal**
- As is disclaimer: Site and content provided as is, no warranty [S5]
- Client controller role: yes · yes [S6]
- Copyright: 2026 · PolyAI Ltd. [S5]
- Copyright holder: 2026 · PolyAI Ltd [S3]
- Entity: PolyAI Ltd. [S4]
- Entity: PolyAI Ltd. · 2026 [S7]
- Entity: PolyAI Ltd [S10]
- Entity: PolyAI Ltd. [S8]
- Governing law: England and Wales [S5]
- Jurisdiction: yes · England and Wales [S5]
- Last updated: 2025-09-15 [S5]
- Notice updated: 9 April 2026 [S6]
- Restriction competing products: may not use Site content to build competing products [S5]
- Trademarks: PolyAI name and logo are trademarks of PolyAI Limited [S5]

**Pricing**
- API credit: Dialog-RSN-1 API · new accounts, limited time · 100 [S8]
- API free credit: 100 [S7]
- Billing granularity: per second of connected time · no · no [S4]
- Free credit: new accounts · 100 · limited time [S1]
- Free credit: new accounts · 100 · yes [S9]
- Free credit: new accounts · 100 · limited time [S10]
- Free credit promo: new accounts · USD · yes · 100 [S3]
- Model: per-minute usage-based pricing [S4]
- Public prices: no public prices; contact via demo form [S4]

**Product**
- 24x7 answering: yes [S2]
- Adk: yes · custom functions · conversation flows · integration with any enterprise system [S9]
- Adk cli: CLI tool · Python package · GitHub workflow [S2]
- Agent build time: enterprise-ready voice agent built in under 10 minutes from plain-language description [S2]
- AI outputs warranty: AI Outputs are informational/illustrative only and may contain inaccuracies [S5]
- API credit: free credit for new accounts, limited time [S4]
- API dialog rsn 1: Dialog-RSN-1 API · yes [S10]
- API free credit: Dialog-RSN-1 · limited time · 100 [S2]
- Audits testing: yes · yes [S3]
- Authentication conversational: Natural-conversation-based authentication, not biometric [S3]
- Automation no integrations: most companies can automate around 30% of all calls with a simple telephony integration, without deep integrations [S7]
- Brand policy adherence: agents follow customer's brand, policies, and rules [S2]
- Channels: voice · chat · messaging [S4]
- Csat claim: 85 [S2]
- Developer docs: ADK Docs available as developer resource [S8]
- Dialog rsn1 API: Dialog-RSN-1 API · yes [S9]
- Dialog studio: Dialog Studio is the named product of the Agentic Dialog Platform [S8]
- Genai gated: Gated generative AI used selectively in conversations for brand safety [S3]
- High availability: yes · yes · yes [S3]
- Industries: Consumer services, Financial services, Healthcare, Hotels, Insurance, Restaurants, Retail, Telecom, Travel, Utilities [S10]
- Industries: Consumer services · Financial services · Healthcare · Hotels · Insurance · Restaurants · Retail · Telecom · Travel · Utilities [S8]
- Labor savings claim: 3 · 10300000 [S2]
- Languages: native-speaker · 75 [S2]
- Llm training data: LLMs trained on millions or billions of conversational samples eliminate the need for extensive training datasets; a couple of hours of call flows is enough to build a deployable assistant within weeks [S7]
- Metrics tracked: Performance tracked via AHT, call volume, and containment; adjustments made to ASR, machine learning, and dialogue design [S7]
- Model upgrades: model upgrades included, no migration project [S4]
- Monitoring: real-time latency and cost per agent monitoring on console [S4]
- Multilingual disputes use cases: fraud · outage · triage · multilingual disputes [S2]
- No code builder: yes [S2]
- Platform capabilities: Account management · Authentication · Call routing · Billing & payments · Booking & reservations · FAQ · Order management · Troubleshooting [S8]
- Proprietary model: Raven · 1B+ enterprise conversations [S2]
- Raven auto reasoning: Hidden reasoning trace for complex requests; simpler turns get direct response [S8]
- Raven latency: outperforms GPT-5 and Claude Sonnet 4.6 across four internal customer service benchmarks · 300 [S8]
- Raven multilingual: Native multilingual thinking, no on-the-fly translation [S8]
- Raven purpose: Raven is purpose-built for live phone calls in voice customer service [S8]
- Raven training: Fused training combining GRPO, DPO, active reinforcement and gold-standard examples [S8]
- Raven version: Raven 3.5 trained exclusively on enterprise customer service conversations [S8]
- Sip header transfer: Custom header information in the SIP header can be transferred back to the contact center to assist agents with handoff [S7]
- Telephony: carrier-grade voice, own carrier over SIP at no charge [S4]
- Testing: test calls, simulated conversations, A/B tests at no extra charge [S4]
- Use cases: Account management, Authentication, Call routing, Billing & payments, Booking & reservations, FAQ, Order management, Troubleshooting [S10]
- Voice AI guide topics: guide covers speech recognition/transcription quality vs resolution rates, context and intent understanding, natural-sounding responses [S10]

**Subprocessors**
- Stripe: Stripe · third party payment processor [S6]

**Support**
- Coverage: 24/7 [S4]
- Demo: demo with live agent · yes · 30 [S10]
- Demo: 30-minute demo offered with a live agent conversation [S8]
- Demo cta: yes [S2]
- Sales contact: Talk to sales [S10]

## Questions

### Is PolyAI European?

Yes, though not from the EU. whats-best.ai classifies PolyAI as UK / wider Europe.

### Where does PolyAI store data?

Not stated on the vendor’s pages as of 26 Oct 2026.

### Does PolyAI have a free plan?

Not stated on the vendor’s pages as of 26 Oct 2026.

### What are the best alternatives to PolyAI?

As of 28 Oct 2026, the panel rates [Langdock (4/5)](https://whats-best.ai/software/ai-agents/langdock), [Dust (3.5/5)](https://whats-best.ai/software/ai-agents/dust) and [Unique AI (3.5/5)](https://whats-best.ai/software/ai-agents/unique-ai) highest among the other 15 AI Agents & Assistants products evaluated. Ratings that look equal are rounded to half stars; the order uses the unrounded rating, then the name. The highest-rated European alternative is [Langdock](https://whats-best.ai/software/ai-agents/langdock) (EU-Made). Ratings are the AI panel’s opinion, formed from public evidence: [how scores are made](https://whats-best.ai/methodology#ordering). The full list, with European products first: [PolyAI alternatives](https://whats-best.ai/software/ai-agents/polyai/alternatives).

## Head-to-head

- [PolyAI vs Parloa](https://whats-best.ai/compare/parloa-vs-polyai.md)

## What this page leaves out

- 18 product facts could not be confirmed on the vendor’s page as captured and were left out of this page and of the panel’s material.
- 4 compliance facts could not be confirmed on the vendor’s page as captured and were left out of this page and of the panel’s material.
- 4 legal facts could not be confirmed on the vendor’s page as captured and were left out of this page and of the panel’s material.
- 2 data facts could not be confirmed on the vendor’s page as captured and were left out of this page and of the panel’s material.
- 2 hosting facts could not be confirmed on the vendor’s page as captured and were left out of this page and of the panel’s material.
- 2 integrations facts could not be confirmed on the vendor’s page as captured and were left out of this page and of the panel’s material.
- 2 sovereignty dimensions could not be confirmed on the vendor’s own pages and are shown as unknown.

## Sources

- [S1] Vendor homepage: https://poly.ai/ (checked 2026-10-26)
- [S2] Features page (agent studio): https://poly.ai/agent-studio (checked 2026-10-26)
- [S3] Security page: https://poly.ai/security (checked 2026-10-26)
- [S4] Pricing page: https://poly.ai/pricing (checked 2026-10-26)
- [S5] Terms naming the legal entity: https://poly.ai/terms-of-use (checked 2026-10-26)
- [S6] Privacy policy: https://poly.ai/privacy-policy (checked 2026-10-26)
- [S7] Models & processing location — found from sitemap: https://poly.ai/conversational-ai/deployment (checked 2026-10-26)
- [S8] Models & processing location — found from sitemap: https://poly.ai/guides/raven-the-llm-custom-made-for-voice-customer-service (checked 2026-10-26)
- [S9] Integrations, connectors & knowledge — found from sitemap: https://poly.ai/integrations (checked 2026-10-26)
- [S10] Agent capabilities & guardrails — found from sitemap: https://poly.ai/guides/the-anatomy-of-an-ai-agent (checked 2026-10-26)

---

Cite as: whats-best.ai, "PolyAI — AI Agents & Assistants", AI panel analysis, as of 2026-10-29. https://whats-best.ai/software/ai-agents/polyai

Corrections: https://whats-best.ai/correct/1035?target=page