Runtime as featured inForbesRead the article

Runtime vs Forethought

Runtime vs Forethought compared: Forethought's multi-agent AI for customer support, now part of Zendesk, vs Runtime's agent harness for fintech support, payment ops, finance, and risk.

Gus Trigos
Co-founder and CEO, Runtime
Updated October 8, 2026 · 7 min read

TL;DR

Forethought is a mature support AI suite, now part of Zendesk, that resolves, triages, and QAs tickets at scale. Runtime fits payment and fintech teams that need agents beyond support, investigating the payment behind the ticket in their own cloud with approvals before money moves.

Feature
RuntimeRuntime
ForethoughtForethought
What it isRuntimeAgent harness for every payment teamForethoughtMulti-agent AI for customer support
Support coverageRuntimeAgents you build from SOPsForethoughtResolve, triage, QA, copilot, insights
Customer-facing channelsRuntimeNot a customer chat widgetForethoughtChat, email, voice, Slack, API
Teams coveredRuntimeSupport, payment ops, finance, risk, complianceForethoughtCustomer support
Helpdesk fitRuntimeConnects to Zendesk and othersForethoughtPart of Zendesk
Where it runsRuntimeYour AWS, GCP, Azure, or self-hostedForethoughtHosted on AWS
ModelsRuntimeAny model, routing and fallbacksForethoughtVendor-managed
Approvals on money movementRuntimeBuilt in, per stepForethoughtNot published
Public pricingRuntimeBy request; free pilot for approved teamsForethoughtNot published

Forethought is the better choice when you want a mature AI suite across your support operation, especially on Zendesk, which acquired it in March 2026. Runtime is the better choice when support tickets lead into payment ops, finance, and risk investigations that need agents of their own, running in your cloud with approvals before money moves. Many fintechs can run both: Forethought on the support queue, Runtime on the investigation behind it.

Forethought and Runtime at a glance

Forethought builds AI agents for customer support and is now part of Zendesk, which announced the acquisition in March 2026. Its multi-agent system covers the support lifecycle: Solve resolves conversations across chat, email, voice, Slack, and headless API; Triage classifies and routes tickets; Discover finds knowledge gaps; a QA agent scores every human ticket; and Assist is a copilot inside the helpdesk. Forethought won TechCrunch Battlefield in 2018 and was supporting more than a billion customer interactions a month by 2025. It is for support organizations, often on Zendesk, that want AI across their whole support operation.

Runtime is the AI agent harness for payment and fintech teams. Support, payment ops, risk, compliance, finance, underwriting, and onboarding each build agents from their SOPs. Agents run on isolated computers in your cloud, reach the ledger, processor, databases, and bank portals, and pause for approval before anything moves money. It is for heads of ops and risk who want agents across the operation, not only in the support queue.

How they differ

A full support suite vs a harness for every team

Forethought goes wide within support. Resolution, triage, routing, knowledge gaps, QA, and agent assist are separate agents built to work together, which is hard to match if support is the only function you are automating.

Runtime goes across teams. Forethought can route a ticket about a missing payout to the right queue. Runtime agents do the investigation behind it, for support, payment ops, finance, and risk on one platform.

Zendesk roadmap vs your own stack

Forethought now sits inside Zendesk. Zendesk has said it will keep supporting existing Forethought customers and fold the technology into its Resolution Platform. For Zendesk shops that is a tighter fit over time. For teams on other helpdesks, it is worth asking how the standalone product will evolve.

Runtime is independent of your helpdesk. Agents reach tools through APIs, databases, CLIs, and MCP servers, with a browser for portals that have no API. Common connections include your processor and ledger, Zendesk, Salesforce, Jira, Snowflake, Postgres, and Datadog.

Where it runs and who holds the data

Forethought runs on AWS infrastructure, holds ISO 27001 and SOC 2 Type II, and is HIPAA, GDPR, and CCPA compliant. It redacts PII, PHI, and financial records during ingestion.

Runtime runs agent computers in your own AWS, GCP, or Azure account, or fully self-hosted with Helm. You bring your own models, including open-weight models in your cloud for PCI and PII work, with routing and fallbacks. Card numbers and SSNs are stripped from prompts and logs, and the record your sponsor bank asks for stays with you.

Money movement and the audit trail

Forethought focuses on resolving and routing conversations. It does not publish a per-step approval model for actions that move money.

Runtime agents start read-only. You choose the steps that need a person, like releasing a payout, applying a reserve, or booking a ledger entry. Every run is recorded end to end, from trigger and tool calls to approvals, cost, and result, and can be exported for a sponsor bank, auditor, or examiner. Solved processes can become deterministic scripts in your repos, owned by your engineers.

Where Runtime enters the picture

Forethought covers the support queue well. The gap is what happens after the ticket is routed. A single stuck payment touches four teams: support gets the "where is my payout" ticket, payment ops traces it through the processor, finance sees a recon break between the ledger and the bank file, and risk or compliance may review the account. Today each team investigates separately in different tools. On Runtime it is one investigation with one record, and support gets the answer back in the ticket.

Runtime is the AI agent harness for payment and fintech teams. You start with one team's sharpest queue, usually support's payment tickets, and grow to every team on the same platform, so the next agent for payment ops or finance starts with the context the first one built.

  • No single vendor to depend on. Your cloud or self-hosted, the sandbox provider you choose with fallbacks, and Claude, GPT, Gemini, or open-weight models with routing across Claude Code, Codex, and OpenCode. Your investigation agents don't ride on your helpdesk vendor's roadmap.
  • Guardrails you set. Read-only by default, approvals before a payout or ledger entry, role-based access, masked credentials, PII and PCI stripping, and every run recorded.
  • Built for teams. Support, payment ops, finance, risk, and compliance share agents, templates, and one memory instead of a tool each.
  • An engineer, not an account executive. A forward-deployed AI engineer with a payments background (Hulu's payments team at Disney, Finix, Modern Treasury, BlackRock) builds the first agents with your team.
New agent

Create a banking ops agent that handles RFIs, hold-harmless requests, and returns of funds from our partner bank, verifies against our ledger, and drafts replies for my approval.

Describe the agent you want
Agent ready

bank-ops-agent

Tools picked from your prompt

EmailLedgerSFTPMastercard
Describe the work in plain English, attach the SOP, and Runtime builds the agent.

Where Forethought is stronger

  • Breadth inside support. Resolution, triage, knowledge insights, QA, and copilot from one vendor.
  • Scale and track record. Years in production and over a billion monthly interactions by 2025.
  • Zendesk alignment. Now part of Zendesk, with integration into its AI products underway.
  • Customer-facing channels. Chat, email, voice, Slack, and headless API.
  • Support analytics. QA on every human ticket and knowledge gap discovery for support leaders.
  • Compliance posture. ISO 27001, SOC 2 Type II, HIPAA, GDPR, and CCPA, with redaction of sensitive data at ingestion.

Pricing

Forethought does not publish prices. Its pricing page describes a blend of platform access fees and an outcome-based cost, across Team, Professional, and Enterprise tiers, sold through sales.

Runtime is available by request, with custom pricing and self-hosting. Approved teams pilot it free on one real workflow first. The value to weigh is the work of analysts you were about to hire across ops, risk, and finance. Rain saved $750k in vendor and salary spend with Runtime, which it uses across technical payment support, fraud and risk transaction monitoring, compliance, cybersecurity, payment operations, engineering investigations, PR reviews, QA testing, and reporting.

Which should you choose

Choose Forethought if

  • You want a full AI suite for support: resolve, triage, QA, and assist
  • You run Zendesk and want AI that is part of it
  • Support is the only team in scope for AI
  • Most tickets resolve from knowledge and helpdesk data

Choose Runtime if

  • You run payments, banking, or lending, and tickets lead into ledger and processor investigations
  • Payment ops, finance, risk, or compliance need agents on the same platform
  • Card data and PII must stay in your cloud, with models you choose
  • You need approvals before money moves and an exportable record of every run

The two can run side by side: Forethought handles the support queue, and Runtime handles the investigation behind it.

Frequently asked questions

Is Forethought part of Zendesk?

Yes. Zendesk announced its acquisition of Forethought in March 2026. Zendesk said it will continue to support Forethought's existing customers and integrate Forethought's technology into its own AI products.

What does Forethought do?

Forethought offers AI agents for customer support: Solve resolves conversations across chat, email, and voice, Triage classifies and routes tickets, Discover finds knowledge gaps, a QA agent scores human tickets, and Assist is a copilot for human agents.

How much does Forethought cost?

Forethought does not publish prices. Its pricing page describes a blend of platform access fees and an outcome-based cost across Team, Professional, and Enterprise tiers, with quotes through sales.

Does Forethought work for fintech?

Forethought lists fintech customers including Acorns, Carta, and YNAB, and fintech use cases like KYC questions and payment tracking. Teams whose tickets require investigation across the ledger, processor, and risk systems may add Runtime behind the support layer.

What is the difference between Forethought and Runtime?

Forethought automates customer support. Runtime is a harness that support, payment ops, finance, risk, and compliance teams use to build agents for internal investigations and operations, running in your cloud with approvals and a full record of every run.

Where does Runtime fit next to Forethought?

Behind it. Forethought resolves and routes the support conversation. Runtime agents investigate the payment behind the ticket across the ledger, processor, and bank files, for support, payment ops, finance, and risk, and hand support a grounded answer with a full record.

See Runtime on your most manual workflow

Bring one SOP. A forward-deployed AI engineer builds the first agent with your team, inside your cloud.