Forethought and Lorikeet are both AI customer support platforms, but they are built for different jobs. Forethought is strongest at triage, agent assist, and augmenting an existing helpdesk: routing tickets, surfacing answers to human agents, and running QA across a support stack. Lorikeet is built for autonomous end-to-end resolution of complex, regulated tickets, with the guardrails and audit trails compliance teams sign off on. The right pick depends on whether you want to make your support operation faster or resolve your hardest tickets without a human in the loop.
Forethought vs Lorikeet is a comparison between two AI customer support platforms that approach support automation from different angles. Forethought is a multi-agent platform built to triage and route tickets, assist human agents in real time, deflect common questions, and score quality across an existing helpdesk. Lorikeet is an AI concierge platform built for complex and regulated industries (fintech, financial services, healthtech, insurance, gaming) where the goal is autonomous end-to-end resolution backed by deterministic and natural-language workflows and a compliance-approvable audit trail. This is a head-to-head on what each is built for, autonomy, regulated-grade guardrails, channels, pricing, and deployment, written to help you shortlist the right one rather than to crown a universal winner.
Forethought's strength is helpdesk augmentation: triage and routing, agent-assist suggestions, knowledge-base deflection, and quality scoring layered onto an existing support stack such as Zendesk, Salesforce, or Freshdesk.
Lorikeet's strength is autonomous resolution of complex, multi-step tickets: deterministic and natural-language workflows in one interaction, defence-in-depth guardrails, omnichannel including sub-1-second voice, and 100% automated QA via its Coach agent.
Pricing models differ. Forethought uses enterprise annual contracts, with publicly reported median total contract values clustering in the tens of thousands of dollars per year and a separate add-on for voice; Lorikeet prices per resolution, roughly $0.80–$0.95 per chat, email, or SMS resolution and about $1.20–$1.50 per voice resolution, with escalations not charged and the customer defining what counts as a resolution.
Both serve enterprise buyers and carry SOC 2. Lorikeet adds BAA-ready (HIPAA) posture, GDPR alignment, PII redaction, RBAC, and US, UK, and AU data residency, which matters for regulated resolution work.
Choose Forethought to automate deflection and make a human agent team faster across an existing helpdesk. Choose Lorikeet when you want to resolve regulated, multi-step tickets autonomously with an audit trail your compliance team can approve before launch.
Last updated: June 2026
Most AI support comparisons assume both products do the same thing and then argue over feature checklists. Forethought and Lorikeet overlap on the surface, both promise to reduce support cost with AI, but their centers of gravity are different, and that is the most important thing to settle before comparing anything else. Forethought is a multi-agent stack designed to sit on top of your helpdesk: it triages and routes incoming tickets, suggests answers to human agents, deflects repetitive questions through self-service, and scores quality afterward. Lorikeet is an autonomous resolution platform: for the tickets it handles, the AI is the agent, and a human is the escalation path, not the default. Both reduce cost and improve quality, but they do it through different mechanisms, and the better fit depends on the shape of your operation. This comparison treats both fairly: Forethought earns real credit for what it does well in triage and augmentation, and Lorikeet is positioned where it genuinely leads, on autonomous regulated resolution.
Forethought vs Lorikeet at a Glance
Forethought · Best for: Support teams that want to triage and route tickets, assist human agents, deflect common questions, and run QA on top of an existing helpdesk · Key strength: Multi-agent triage, routing, agent assist, and quality scoring across a support stack · Primary mode: Helpdesk augmentation and deflection, with autonomous resolution of common tickets · Pricing: Enterprise annual contracts, with a separate voice add-on · Compliance: SOC 2
Lorikeet · Best for: Complex and regulated companies (fintech, financial services, healthtech, insurance, gaming) needing autonomous end-to-end resolution with audit trails · Key strength: Resolution depth on multi-step regulated tickets; defence-in-depth guardrails; 100% automated QA · Primary mode: Autonomous resolution (AI is the agent), with human escalation · Channels: Chat, email, voice (sub-1-second latency), SMS, WhatsApp, plus outbound re-engagement · Pricing: Per resolution (~$0.80–$0.95 chat/email/SMS, ~$1.20–$1.50 voice; escalations not charged) · Compliance: SOC 2, BAA-ready (HIPAA), GDPR-aligned, US/UK/AU data residency
What Each Platform Is Built For
Forethought is a multi-agent AI platform built to make an existing support operation faster and more efficient. Its agents cover the support workflow end to end on the operations side: deflecting and resolving common questions through self-service, triaging and routing incoming tickets to the right queue or agent, assisting human agents in real time with suggested answers and relevant context, surfacing gaps in knowledge and process, and scoring quality across interactions. Forethought is strong for teams that already run a helpdesk such as Zendesk, Salesforce, or Freshdesk and want to layer intelligence onto it: cut handle time, route more accurately, deflect the repetitive tickets, and lift the performance of the human agents who handle the rest. Following its acquisition by Zendesk announced in March 2026, that helpdesk-native positioning becomes even more central.
Lorikeet is an AI concierge platform built specifically for complex and regulated industries. Roughly 80% of its customers are US financial institutions and fintechs. Rather than triaging tickets to humans or assisting them, Lorikeet builds concierges that resolve issues end-to-end across channels, and a second agent, Coach, that runs analytics and 100% automated QA. The design center is the hard ticket, the kind that needs several actions executed in the right order against live systems, with a record a compliance team can replay. Where Forethought optimizes for routing and augmenting the support stack you already run, Lorikeet optimizes for autonomously resolving the ticket so it never needs to be triaged or handed to a human at all, unless a guardrail or your policy says it should.
Autonomy: Triage and Assist vs End-to-End Resolution
The clearest difference between the two platforms is what happens to a ticket once it arrives. With Forethought, much of the value sits before and around the human agent: an incoming ticket is classified and routed to the right place, common questions are deflected through self-service, and when a human does pick up a ticket, the AI suggests answers and surfaces context to speed them up. Forethought also resolves a meaningful share of routine tickets autonomously through its Solve agent. That model is genuinely valuable. Better triage reduces misroutes and reopens, deflection removes repetitive volume, and agent assist shortens handle time and ramp for new hires. For a support organization whose strategy is a more efficient helpdesk with faster, better-supported human agents, that combination is the right mechanism.
With Lorikeet, the AI is the agent for the tickets it handles, and the design point is the complex ticket rather than the common one. A question like "why was my transfer declined, and can you refund the fee" is not routed to a human or answered from a knowledge article; it is resolved by the concierge verifying identity, checking the transaction in a payment system, applying policy, taking the action, and confirming the outcome, keeping state across all of it and recovering gracefully if a step errors. Lorikeet combines deterministic structured workflows with natural-language workflows, and the two can run in a single interaction, so a flow that must follow an exact regulated script can hand off to flexible reasoning and back without leaving the conversation. The Team of Agents capability dispatches sub-agents to call third parties and coordinate, for example contacting a merchant on a dispute or a pharmacy on a prescription question. The honest tradeoff: autonomous resolution of regulated work is more configuration than turning on triage and assist, which is why Lorikeet ships with forward-deployed implementation help rather than expecting you to self-serve the hardest workflows alone.
Regulated Guardrails
For a regulated buyer evaluating autonomous resolution, the question is not "does it route and assist well" but "can my compliance team approve the agent acting on its own before launch." That is where Lorikeet concentrates its design effort, and it is a question that matters differently for an autonomous agent than for a triage and assist tool. When a human still picks up the ticket, that human is a control; when the AI resolves end-to-end, the controls have to be built into the system.
Lorikeet runs what it calls defence in depth: pre-launch adversarial simulations and red-teaming, inbound message checks, outbound guardrails, and 100% post-facto QA through the Coach agent. The framing the team uses is that the large language model is the engine and Lorikeet is the cockpit. In practice that means you can simulate the bad paths before go-live, read the results, and prove the agent declines to act when a guardrail trips, rather than discovering the failure mode in production. These features support your compliance obligations; they do not by themselves certify compliance, and any vendor that promises certification is overstating it.
Forethought operates with SOC 2 compliance and provides controls suited to its model: routing rules, confidence thresholds that decide when to deflect versus escalate, and quality scoring through its QA agent that reviews interactions after the fact. That is a real strength for a triage and augmentation platform, where the human agent and the helpdesk's own controls remain part of the loop. The distinction is the unit being governed: Forethought's controls shape how tickets are classified, deflected, and reviewed within a human-supported stack, while Lorikeet's are designed to govern an autonomous agent's actions end-to-end, with simulation-based validation before launch and automated QA on every ticket afterward. If you are augmenting a helpdesk, Forethought's routing and review controls fit the model; if you are resolving autonomously in a regulated domain, the deeper requirement is to prove the agent behaves correctly without a person there to catch it.
The reason this matters is the cost of the tail. Any platform can post strong aggregate numbers while still failing on the small set of interactions that carry real risk, a disclosure left out of a collections message, a PII detail surfaced to the wrong party, an action taken above a dollar threshold that should have required human approval. Aggregate deflection and resolution metrics hide those cases by design. Lorikeet's approach is to make the tail testable: you can run adversarial simulations against the exact scenarios that worry your compliance team, read a pass-or-fail report, and gate launch on it, then keep Coach watching every ticket afterward so a regression shows up in QA rather than in a regulator inquiry. That posture is specifically built for letting an AI act on its own in regulated work.
Channels
Forethought supports the channels common to a modern helpdesk, including chat, email, and messaging, and offers voice through a separate add-on, which fits a platform whose core is triage, deflection, and agent assist across a support stack. For an operation centered on an existing helpdesk, that breadth covers the channels most teams already run.
Lorikeet covers chat, email, voice, SMS, and WhatsApp on the inbound side, plus outbound re-engagement over voice, SMS, and email for use cases like collections and abandonment, with compliance controls for do-not-call lists, call-hour rules, and consent. Its voice agent runs at sub-1-second latency with natural conversation and automatic language switching. The architectural point that matters for autonomous support is that voice runs on the same workflow engine as chat and email, so the agent carries shared context across channels and can take actions on a call rather than routing to a human. A customer who starts in chat and moves to voice does not have to repeat themselves, and the agent can lock a card or file a dispute live. The difference from Forethought is again about who acts: Forethought's channels feed a stack where humans and the helpdesk remain in the loop, while Lorikeet's voice agent handles the call itself.
Pricing
The platforms use different commercial models, and the difference follows directly from the augmentation-versus-autonomy split.
Forethought uses enterprise annual contracts. Publicly reported third-party data puts median total contract values in the tens of thousands of dollars per year, with a range up into the low hundreds of thousands depending on volume and agent count, and voice is typically a separate add-on. Exact figures are not published by the vendor, so the right reference point is a custom enterprise quote based on your ticket volume, agent count, and which of the agents in the stack you turn on. For a team already running a helpdesk, that annual-contract shape is a familiar way to budget.
Lorikeet prices per resolution and lets the customer define what counts as a resolution. Chat, email, and SMS resolutions run about $0.80–$0.95 each, voice resolutions about $1.20–$1.50, and the Coach QA agent runs about $0.25–$0.30 per ticket and can be deployed standalone. Escalations to a human are not charged, which removes the incentive for the vendor to claim a resolution it did not earn. Set against a human baseline of roughly $1.25 to $4 per handled ticket, per-resolution pricing is designed to track value rather than seats or volume tiers.
The practical question for a buyer is which model maps to your strategy. If you are keeping a helpdesk and a human agent team and want to route better, deflect more, and assist faster, an annual-contract platform like Forethought charges in proportion to the operation you are equipping, which is a clean fit. If your goal is to remove complex tickets from human queues entirely, per-resolution pricing charges only for the work the AI actually completes, and because escalations are not billed and the customer defines a resolution, you are not paying for interactions the agent did not finish. The two models are not really competing on price so much as on what you are buying: a more efficient helpdesk, or tickets resolved without an agent.
Deployment
Forethought deploys onto an existing helpdesk and integrates with major platforms such as Zendesk, Salesforce, and Freshdesk, then learns from your historical tickets and knowledge base to tune deflection, routing, and assist to your domain. The deployment effort centers on connecting the helpdesk, ingesting knowledge and ticket history, configuring routing and confidence rules, and rolling assist out to a human agent population. For organizations that already run a support stack and want to layer intelligence onto it, that is a familiar shape of project, and the recent Zendesk acquisition reinforces that helpdesk-first integration story.
Lorikeet pairs a plain-English configuration model, where workflows, guardrails, and tools are defined in natural language, with a forward-deployed product manager and engineer during implementation. A working sandbox is typically standing in 20 to 30 minutes, with a production deployment operational in about a month. The tradeoff is honest: standing up autonomous resolution of regulated flows asks for more upfront partnership than enabling triage and assist on an existing helpdesk, because the agent has to do the whole job correctly rather than route or suggest it. The payoff is that the hard, regulated flows are built and validated before they go live.
Where Forethought Is the Stronger Choice
It would be unfair to frame this comparison as if autonomy were always the goal. For many operations it is not, and Forethought is the better fit for several real situations.
If your strategy is to get more out of an existing helpdesk and human agent team, Forethought is built for exactly that. Its triage and routing reduce misroutes and reopens, its deflection removes repetitive volume before it reaches an agent, and its real-time assist shortens handle time and new-hire ramp while making quality more consistent. Its QA agent reviews interactions across the stack instead of sampling a handful, turning quality management into something continuous. For teams already committed to Zendesk, Salesforce, or Freshdesk, especially after the Zendesk acquisition, Forethought slots into that ecosystem rather than asking you to change helpdesks. And if a large share of your volume is genuinely common, repetitive questions plus tickets that a human should still own, augmentation captures the efficiency gains without requiring you to fully automate. The honest summary is that Forethought wins when the helpdesk and a human agent team are central to your strategy and the goal is to make that operation faster and more accurate, and for a great many support teams that is exactly the right goal.
Feature-by-Feature Summary
Primary mode: Forethought augments a helpdesk with triage, routing, deflection, agent assist, and QA, and resolves common tickets; Lorikeet resolves complex tickets autonomously end-to-end, with humans as the escalation path.
Resolution depth: Forethought routes and deflects, surfaces answers to a human, and autonomously resolves common questions; Lorikeet executes multi-step regulated tickets itself using deterministic plus natural-language workflows in one interaction.
Regulated guardrails: Forethought provides routing rules, confidence thresholds, and post-hoc QA with SOC 2 coverage; Lorikeet adds defence in depth (pre-launch simulations, message checks, outbound guardrails, 100% QA) designed to govern an autonomous agent before go-live.
Channels: Forethought covers chat, email, and messaging with voice as an add-on; Lorikeet runs chat, email, voice, SMS, and WhatsApp plus outbound, with voice on the same engine at sub-1-second latency.
Pricing: Forethought uses enterprise annual contracts with a separate voice add-on; Lorikeet prices per resolution (~$0.80–$0.95 chat/email/SMS, ~$1.20–$1.50 voice) with escalations not charged.
Compliance posture: Both SOC 2; Lorikeet adds BAA-ready (HIPAA), GDPR alignment, PII redaction, RBAC, and US, UK, and AU data residency.
How to Choose Between Forethought and Lorikeet
The decision comes down to whether you are making a helpdesk more efficient or removing complex tickets from human queues entirely.
Choose Forethought if your strategy is built around an existing helpdesk and a human agent team you want to make faster and more accurate through triage, routing, deflection, agent assist, and QA. Forethought's multi-agent stack is a real strength, and for teams committed to Zendesk, Salesforce, or Freshdesk that want to lift the performance of their support operation, it is exactly what the job needs, particularly given its tight Zendesk integration after the acquisition.
Choose Lorikeet if you operate in a regulated or complex domain, your important tickets require several actions in the right order against live systems, and you want to resolve them autonomously with behavior your compliance team can approve before launch. Lorikeet leads on autonomous resolution depth, defence-in-depth guardrails, deterministic plus natural-language workflows, single-engine omnichannel including sub-1-second voice, simulation-based validation, audit trails, and 100% automated QA. The cost of that depth is a more involved implementation, which is why Lorikeet provides forward-deployed help rather than leaving you to self-serve the hardest flows. The two can also coexist: triage and assist on the common tickets that a human should own, and resolve the hard regulated ones autonomously.
If your hardest tickets are regulated and multi-step, book a Lorikeet demo and bring your toughest 10 tickets. We will run them in your stack against your guardrails before you sign.
Key Takeaways
Forethought and Lorikeet solve the support problem from different angles: Forethought augments a helpdesk with triage, routing, deflection, agent assist, and QA, while Lorikeet resolves complex regulated tickets autonomously end-to-end.
On autonomy, Forethought routes, deflects, and assists within a human-supported stack and resolves common tickets; Lorikeet makes the AI the agent for hard tickets, combining deterministic and natural-language workflows and dispatching sub-agents for multi-step work.
On guardrails, Forethought provides routing rules, confidence thresholds, and post-hoc QA with SOC 2 coverage; Lorikeet's defence in depth (pre-launch simulations, message checks, outbound guardrails, 100% QA) is built to govern an autonomous agent before go-live.
Pricing follows the model: Forethought is enterprise annual contracts with a separate voice add-on, Lorikeet is per resolution (~$0.80–$0.95 chat/email/SMS, ~$1.20–$1.50 voice) with escalations not charged.
Pick by your strategy: making an existing helpdesk faster points to Forethought; autonomously resolving regulated, multi-step tickets points to Lorikeet.
