Proto offers ticketing metrics within the Analytics module, available on all plans.

Yes – you can train AI agents on your content, build structured workflows, configure languages and deployment channels, and tailor tone persona of your agent using prompts.

Yes, your data will be automatically deleted after 30 days if you request account deletion. You can export your data directly from your Proto account or request support from our team.

Yes – you can configure individual sources across Facebook, X (Twitter), Instagram, YouTube, TikTok, news websites, Google Play, and the Apple App Store. Each source can be enabled or disabled independently.

Yes – you can downgrade your plan in the platform.

Yes – you can sync AI agents with URLs and upload documents (PDFs, tables, JSON, and transcripts). Training sources can be added, edited, or removed at any time.

Yes – you can review AI agent conversations in the Inbox. Use it to identify gaps where the agent lacked the right source content, and improve performance over time with better training data and workflows.

Yes – you can customise instructions for your LLM–powered AI agent, and use intents, triggers, and structured flows for specific journeys (e.g., refunds, complaint intake, bookings). AI agents can also call APIs using your authentication rules.

If you opt for the annual payment option, you can still use additional interactions as needed on a monthly basis. If you go over your interaction limit, the payment will be reconciled.

Yes. Proto engages with organisations whose capabilities strengthen inclusive, secure, and scalable AI deployments in regulated environments. Please express your interest by submitting an application form.

Yes, you can use Proto’s APIs to send information collected in Proto to your sales and marketing platforms.

Yes. On on-premise deployments, Proto can run locally hosted models – including DeepSeek V3, Llama and Qwen – so no inference leaves your infrastructure.

Third-party and dedicated LLM options are covered in the full FAQ.

Yes – AI agents can use your knowledge base (with LLM enabled) to answer complex questions, while escalating sensitive or unresolved cases to live agents. You can also build manual workflows for structured customer journeys (e.g., for refunds, complaint intake, bookings). Workflows provide full control over building a complex agentic functionality.

Yes – monitor conversations in the Inbox, test the AI agent before deployment, add or refine training sources, and adjust prompts or workflows to improve how the AI agent responds over time.

Yes, you can sync Proto with your CRM so all persons reflect in your CRM and/or Proto. You can use this data in your AI agent workflows for personalised conversations by using custom variables in the AI actions.

Yes – Proto can integrate with any system that supports API. You can connect CRMs, ERPs, data stores, and internal tools to power workflows inside AI agents, Inbox, and Tickets.

Yes, Proto offers APIs that can be used to pull or push information from and to Proto to other systems.

No – unused additional credit bundles expire at the end of the billing period. Consider upgrading your plan if you see a constant credit overage.

No – interaction limits reset each month. If you need more in a given month, you can add additional interaction bundles.

Exclusivity is determined by the partner’s claimed deal volume and minimum revenue guarantee. Exclusivity may be provided for either a country, region, and/or industry.

Proto has a Starter plan that is completely free to use allowing you to use up to 1,000 interactions per month. You can sign up for a free Starter plan now in a few minutes. Larger businesses, enterprise and government agencies that wish to try unlimited platform features before subscribing can contact our team.

Yes – API access is available on specific plans. See the Developer API for details.

When a partner joins the Proto partner network, Proto publishes an announcement on our website and social media. Based on the partner’s preferences, larger announcements, events, and long-term marketing can be performed.

Yes, as a requirement of SOC2 and ISO:27001 standards, Proto conducts penetration testing once a year.

Yes – you can build AI agents using custom structured conversation workflows (for precision and compliance) and generative responses powered by the selected LLM (for flexibility). Workflows that our clients build usually use a mix of both.

Yes – Proto is backed by audited security standards including SOC 2 Type II and ISO 27001. The platform also supports HIPAA and GDPR requirements for regulated workflows.

Learn more: Security and Trust Centre.

Yes, all chats with live agents are available in the module’s history tab. Key analytics are available in the Analytics module.

User data is stored on our cloud servers, which are hosted by AWS. We keep client data for the duration of your platform subscription, and it is removed 30 days after the subscription ends, allowing you the chance to export any necessary data from the platform. If requested, your data can be deleted before this period.

You can configure alerts to fire when a keyword is detected, when mention volumes reach a set threshold within a chosen time window, or when the share of negative sentiment exceeds a defined percentage. Alerts are sent by email and can be filtered by entity, channel, sentiment, or category.

Chats can be handed off from AI agents to live agents automatically (based on any configured condition) or a live agent can take it over in the Inbox manually. Routing options include round–robin, load–balanced, and manual assignment.

Learn more: Livechat and Inbox.

Technology partners integrate directly into Proto’s AI agent infrastructure – contributing capabilities such as fraud detection, extended analytics, payments, or network intelligence that enhance supervisory and operational workflows.

To cancel your Enterprise plan, raise a ticket with the Proto support team or inform your assigned CX Manager.

By combining detection tools, analytics, and AI-powered complaint workflows, partnerships enable end-to-end traceability – from risk signals to investigation, enforcement, and measurable resolution outcomes.

You define keyword groups – a primary keyword and its variations – and Perception only stores entries that match. This keeps your data focused and avoids noise from unrelated content.

Scraped and analysed data feeds directly into Proto’s Analytics module, where it can be compared with livechat and ticketing data. This lets you correlate public sentiment trends with operational CX metrics in a single view.

When you cross the monthly interactions limit, you start using additional interactions. At the end of the billing month, your selected payment method will be charged for the additional interactions. If you are on an annual plan, the additional interactions will be charged on the annual renewal or reconciled during the year.

Proto detects language shifts mid–conversation and responds in the dominant or requested language, including mixed language patterns such as Taglish.

Learn more: Local languages.

Proto prioritises data security and confidentiality by implementing robust encryption protocols, access controls, and biannual penetration testing. Additionally, sensitive data is anonymised or pseudonymised whenever possible to minimise risks. Access to confidential information is strictly limited to authorised personnel, and all interactions with the platform are logged and monitored to detect and prevent any unauthorised access or data breaches.

Every message goes through one of the configured workflows. LLM-generated response can be triggered by a custom LLM action in any of your workflows or when fallback is triggered.

For response generation, you have to enable third–party LLMs (available on any plan), use a dedicated LLM (Embed), or self–host your own model (available for Enterprise on–premise deployments).

The AI agents secure API interactions with encryption protocols like HTTPS to protect data integrity and confidentiality in transit. For authentication, Proto uses API keys and OAuth tokens to guarantee that only authorised users access these APIs. To maintain data integrity, Proto enforces strict input validation and sanitisation practices on enduser inputs prior to processing.

Proto also applies rate limiting and throttling to prevent misuse and promote equitable resource use. Through regular security evaluations and vulnerability scanning, Proto proactively identifies and mitigates potential security threats.

Adhering to OWASP recommendations and industry norms for API security, along with continuous activity monitoring and logging, Proto swiftly addresses any abnormal actions. This approach underscores Proto’s dedication to ensuring a secure, trustworthy exchange with external systems, protecting user data’s privacy and integrity.

When you pay annually, your credit limit is still reset monthly. Every month, starting on the date when you purchased the plan, your base credit volume will be reset.

The information is secured both when stored and during transmission. The security protocols employed include TLS and SSL.

Plans are priced based on included features and monthly credit volume. If you reach your monthly limit, you can add additional credit bundles.

Learn more: Pricing.

Enterprise pricing is typically based on monthly interaction volume and required service scope (e.g., hosting model, integrations, and support). Enterprise plans include access to all platform features, with custom terms available for higher volumes.

Learn more: Enterprise and Pricing.

Additional credits remain available until the end of your current billing period. Unused credits do not roll over to the next period.

Proto is backed by philanthropy partners such as the Gates Foundation to deploy capital via the shared service model at an initial six (6) countries in Southeast Asia and Africa. Exact funding per country depends upon the number of government agencies within a deployment. The funding pays 100% of a one (1) year license fee for the Proto platform and solution design by the Cambridge SupTech Lab.

No – you can start using Proto without entering card details. You will only be asked for payment details if you upgrade to a paid plan.

Learn more: Pricing.

Yes – Proto is an official WhatsApp Business Tech Provider and a Meta Business Partner. This means businesses using Proto’s WhatsApp channel benefit from a direct, verified integration with WhatsApp’s platform.

Yes, you can set up a feedback collect action and see your CSAT performance in the Analytics.

Limits depend on your plan. Some items (like tickets and livechats) are not limited, while other limits (like number of AI agents or languages) vary by tier.

Learn more: Pricing.

Yes – voice messaging (speech–to–text and text–to–speech) is included across all plans for both common and local languages.

Proto’s AI agents can function with or without enabled third-party LLMs.

Without enabled LLM, functionality will be limited to workflows (knowledge base will not be available). This setup provides extended data control, as data isn’t shared with external providers. AI agent will function using Proto’s own AI engine.

For response generation, you have to enable third–party LLMs (available on any plan), use a dedicated LLM (Embed), or self–host your own model (available for Enterprise on–premise deployments). Our Enterprise solution provides the most extended data control options.

Proto AI agents can be configured to use third–party LLMs, a dedicated LLM, or a self–hosted model (including on–premise). In parallel, Proto provides localisation for underserved and mixed languages, plus controls for what data is shared with external providers.

Learn more: AI Agents, Data privacy, and Local languages.

The Proto platform offers two primary levels of user roles: Workspace and Teamspace. Each level is customisable across a comprehensive set of approximately 50 permissions. This granularity allows for precise control over user access and features, enabling organisations to tailor the security of the platform to meet their specific needs.

Proto’s system is equipped with extensive logging and tracking capabilities to monitor user activities. This includes keeping a detailed record of modifications in platform settings, assistant configurations, ticketing, and live chat details, providing a clear audit trail of all changes.

Low-to-middle income countries can qualify for the shared service model if they intend to integrate AI agents and case management across >2 government agencies with a mandate for consumer protection and/or regulation. An example of this would be a central bank, insurance commission, and securities exchange seeking to share a consumer protection system.

Tickets can be created automatically by your AI agents within configured workflows, or by your live agents. Live agent can create a ticket from scratch, or make one from an existing livechat. You can also connect your team’s mail boxes so tickets are created automatically whenever you receive an inbound email.

After the one (1) year of funding from philanthropy partners, the government agencies can commit to split the ongoing license fee for the Proto platform at a reduced per-agency cost. Alternatively, the agencies can develop the technology further and independently under the philanthropy partners’ license. During the deployment, government agencies are expected to participate in the provision of citizen case / complaint data, technical implementation, user acceptance testing, and cross-agency collaboration. A project charter defines the responsibilities of government agencies in detail.

Proactive messaging is available on Webchat. You can engage your website visitors with a proactive popup message triggered by custom page-based rules or API.

When a customer sends a voice message, Proto transcribes it into text. Live agents can read the transcript and play the original audio in the Inbox.

Replies can be delivered to the customer as both text and voice (if channel support audi messages) in the selected language and accent.

Each scraped entry is processed as two interactions – one for the scrape and one for the AI analysis (categorisation and sentiment). If analysis fails and is re-triggered manually, no additional interaction is charged.

If your payment method is removed or a transaction fails, outbound replies stop. This means AI agents will not respond and live agents will not be able to send messages.

Inbound messages from customers will still be received, and you can resume replying once the billing issue is resolved.

If you go over your base interactions volume consistently, Proto will suggest upgrading your base volume. This will be available at the base volume price per interaction, leading to a reduction in cost when compared with paying for additional interactions every month.

Proto specialises in the inclusion of historically-marginalised citizens, with text and voice AI capability for local languages in Africa and Asia. These localised AI agents are available across SMS for feature phones and common messaging apps like WhatsApp and Messenger.

The shared service model is designed to enable deployment with one (1) year of philanthropy funding. This funding allows for the generation of citizen experience and social impact metrics to support complex procurements. Additionally, the shared service model allows participating government agencies to split the license fee for the Proto platform amongst themselves. Typically, the final cost per-agency falls below small-value procurement levels or within budgetary limits.

Please apply for funding. In this case, Proto and its partners may be able to assist in outreach to other agencies in your country in order to enable the shared service model.

A workflow is a set of actions an AI agent executes when a specific user intent or event is detected.

Workflows can be triggered by events such as:

• Message received – when a user sends a new message
• Attachment received – when a user sends a file
• Fallback – when the AI agent cannot confidently match the message to a known workflow

Every AI agent also includes a baseline set of default workflows to handle key lifecycle events, such as Chat started, Chat closed, Chat timeout, AI timeout, and Fallback.

Proto prioritises rapid response to critical incidents. Information about any possible disruptions and downtime is relayed through Proto’s status page at status.proto.cx.

You can cancel your subscription in the platform. For accountdeletion , please contact the Proto support team.

If you are on a paid plan, your subscription will remain active until the end of the current billing cycle.

Proto and its implementation partner, the University of Cambridge SupTech Lab, conduct a twelve (12) month comprehensive solution design, implementation, testing, and deployment. This process provides at least three (3) months for production-level deployment to gather citizen experience and social impact metrics to support continued procurement.

Technology partners contribute technical capabilities or integration expertise.
Development partners support national or multi-country deployments through funding, ecosystem coordination, or programme design.

You can submit your application on the Proto website. The relevant region manager will review the application. If approved, you will receive a Partner Agreement for signing.

Proto is constantly maintaining its integrations with these third-party apps. You can also ask the assistant in the Proto platform if you need additional help with enabling these apps.

Partners are given access to the Proto Partner Portal, containing sales materials, FAQ, and documentation. Partners can receive training sessions from prompt engineers and sales demo assistance from region managers.

Proto supports 100+ languages, including underserved dialects and mixed languages (e.g., Taglish), with automatic mid–conversation language detection and switching.

Proto supports voice messaging across common global languages and selected underserved local languages, with a focus on Southeast Asia and Africa (e.g., Tagalog, Kinyarwanda Oshiwambo).

Learn more: Local languages.

Proto uses Security Journey to offer ongoing security training to its employees and comply with SOC2 standards. This method guarantees that employees are consistently informed about potential security challenges and follow established guidelines for best practices in security.

Proto incorporates comprehensive business continuity and disaster recovery protocols to guarantee service availability at all times. This includes routine backups of vital data and system configurations, alongside redundant components to reduce the risk of singular failure points. Moreover, the infrastructure supporting Proto’s service is strategically distributed across multiple geographic data centers, ensuring the availability of failover options during unforeseen incidents. Proto also regularly undertakes tests and simulations to affirm the effectiveness of its disaster recovery plans, aiming for swift and efficient restoration of services in the event of any disruptions.

The system uses email and password-based authentication, as well as Single Sign-On (SSO) options through Google and Microsoft, and it includes support for multi-factor authentication.

Proto accepts credit card payments for self–serve plans. Alternative payment methods may be available for Enterprise and government clients.

Proto accepts credit card payments for all plans. Wire transfer payments are available for Enterprise clients.

Proto consistently conducts reviews mandated by SOC2 and ISO:27001 standards, such as biannual penetration tests, quarterly vulnerability scans, and security assessments, to identify and evaluate security risks. Additionally, internal security tests are carried out, and proactive steps are taken to strengthen the security of our solutions.

Proto’s reseller program has three tiers - referral, collaborative, and autonomous - depending on a partner’s engagement level with clients. Revenue varies accordingly up to 30%.

The Proto platform offers granular roles and permission levels that are fully customisable to your needs. You can create roles and define precisely what each team member can access and at which level.

Proto supports data privacy requirements such as GDPR and can be deployed for HIPAA–aligned workflows. Data is protected behind robust firewalls on secured infrastructure, with controls such as IP allowlisting, SSO, and multi–factor authentication.

Learn more: Data privacy and Security.

Clients onboarded by partners receive the same client experience as direct-sale clients. Partners are trained to manage and resolve client queries, and Proto’s prompt engineers are standing by to provide second-tier technical assistance.

Proto collaborates with technology partners, development institutions, digital public infrastructure initiatives, and strategic ecosystem organisations that strengthen regulatory-grade AI deployments.

If you have additional credits enabled, a bundle can be applied automatically as you reach your monthly limit to avoid disruption.

If additional credits are not enabled, outbound replies pause until you purchase more credits or your next billing cycle starts.

Learn more: Pricing.

Proto and client data can be hosted on Amazon Web Services (AWS) in supported regions, including Singapore and Europe.

For data sovereignty needs, Proto supports hybrid and on–premise hosting, including hosting chats, people, and tickets in your own databases (available on Enterprise plans).

Learn more: Data privacy and Enterprise.

Government agencies own the data within their accounts under the shared service model, and can export it at any time.

Proto supports cloud, hybrid, and on–premise hosting options to meet data privacy and sovereignty requirements, including dedicated or self–hosted language models when needed.

Learn more: Data privacy and Security.

Proto builds AI systems as part of broader digital public infrastructure. Interoperability ensures that detection, complaints, supervision, and enforcement systems work together rather than operating in silos.