Most enterprise AI vendors sell a vision. Aisera sells a ticket count. The company was founded in 2017 in Palo Alto by Muddu Sudhakar and Christos Tryfonas, and its core promise hasn’t shifted much since: take the routine, repetitive requests that clog IT service desks, HR help centers, and customer support queues, and resolve the majority of them without a human touching them.
That narrowness is why Aisera AI keeps surfacing in CIO conversations even though it rarely generates the media noise of the big model labs. It isn’t trying to build the smartest model on earth. It’s trying to wire capable models into the systems that already run a business, and collect a per-seat fee for the privilege.
What Aisera AI actually is
Strip away the product names and Aisera is a workflow engine with a conversational front end bolted on. An employee opens Slack, Teams, a web portal, or simply emails the help desk and asks something that has been asked ten thousand times before. A password reset. A VPN that won’t connect. A request for a replacement laptop. A question about parental leave policy.
An AI agent reads the request, pulls context from the company’s knowledge base and connected systems, then either answers directly or executes the underlying action. Resetting a credential in Okta. Logging a ticket in ServiceNow. Checking remaining PTO in Workday. Approving a software license. If the agent hits something it can’t resolve, it hands the conversation to a person with a summary and a draft reply already written.
The customer list includes names like Autodesk, Zoom, and McAfee. Aisera’s published benchmarks claim up to 80% of employee requests resolved autonomously, along with steep drops in average handle time. Those come from the vendor’s own customer results, so treat them the way you’d treat any supplier’s numbers: directionally useful, not independently audited.
Inside the model stack
AiseraGPT is the label the company gives its domain-tuned layer. Underneath, the platform is deliberately model-agnostic. It can call OpenAI, Azure OpenAI, Anthropic, Gemini, or open-weight options like Llama, and it also ships smaller fine-tuned models for narrow tasks where latency and cost matter more than raw reasoning power.
Two design choices carry most of the weight:
- Retrieval grounded in the customer’s own content. Answers are pulled from ServiceNow knowledge articles, Confluence pages, SharePoint documents, and previously resolved tickets. That grounding is what keeps hallucination rates survivable in a regulated environment.
- Confidence thresholds and hard escalation rules. Agents act on their own only when a confidence score clears a bar the customer sets. Everything else routes to a human, and every interaction is logged for audit.
Deployment flexibility matters more than most buyers expect. Aisera runs as multi-tenant SaaS, inside a customer’s own cloud tenancy, or on-premises for organizations that can’t send employee data to a third party. It markets SOC 2 Type II compliance alongside support for HIPAA and GDPR workloads, which is the price of entry for healthcare and financial services deals.
Where it lands inside a company
IT service desk
This is the beachhead, and it’s where the volume lives. Password and access issues, VPN trouble, laptop provisioning, and software requests typically make up 40% to 60% of tier-1 tickets at a mid-sized company. Aisera’s pitch is that an agent resolves those in seconds, at any hour, in whatever language the employee writes in. That’s also the story behind how Aisera is rebuilding the service desk around AI agents rather than around ticket queues.
HR and employee services
HR ticket volume is smaller but messier, full of policy nuance and sensitive data. Benefits enrollment, payroll questions, onboarding checklists, and leave requests all live here. The value isn’t headcount reduction so much as giving a five-person HR operations team the capacity of fifteen during open enrollment.
Customer-facing support
Harder territory. External customers are less forgiving than employees, and a wrong answer has a revenue cost attached. Billing questions, order status, subscription changes, and returns are the usual starting points. Here Aisera runs into Intercom’s Fin, Sierra, and Salesforce Agentforce, all chasing the same containment numbers.
The competitive squeeze
Moveworks was Aisera’s closest analog until ServiceNow agreed to acquire it for $2.85 billion in March 2025. That deal validated the category and simultaneously made life harder for independents, because the largest system of record now ships its own AI agents at no extra line item.
Aisera’s counter is neutrality. It works across ServiceNow, Salesforce, Zendesk, Jira, Workday, and Slack rather than privileging one. For a company that has spent a decade avoiding vendor lock-in, that’s a genuine argument. For a company standardized entirely on ServiceNow, it’s a much weaker one.
What to scrutinize before signing
Pricing isn’t published, and enterprise contracts commonly land in the six-figure range once implementation and professional services are included. Before you get that far, get straight answers on a few things:
- Containment versus deflection. Ask what percentage of conversations end without a human, not how many tickets the system “avoided.” The two numbers are often wildly different.
- Knowledge base hygiene. Agents are only as good as the articles behind them. If your top 50 help articles contradict each other, no model will save you.
- Cost per resolution. Take your fully loaded tier-1 cost, usually $15 to $30 per ticket, and model what a 50% containment rate actually saves after licensing.
- Data handling. Confirm whether your content is used to train shared models, where data resides, and how long it’s retained.
- Pilot scope. Start with the top 20 request types for one department. Broad rollouts hide the fact that the agent only performs well on a narrow slice of intents.
The math that decides it
A 40,000-ticket service desk burning $22 per tier-1 resolution spends roughly $880,000 a year on work that is largely repetitive. Contain 55% of it and the savings are real, provided the platform doesn’t simply shift cost into knowledge management, integration work, and the ongoing job of keeping content accurate.
That last part is the honest catch with any agentic support platform, Aisera included. The AI is the easy half. The hard half is the unglamorous process mapping and article cleanup that has to happen before an agent can be trusted with the keys. Teams that budget for both tend to get the numbers the vendor promised. Teams that buy the software and hope the messy documentation sorts itself out usually end up with a very expensive chatbot.