Squidcom
In 2026, AI support agents have become essential to good customer service. A few years back, chatbots were clunky and frustrating. But today's large language models deliver accurate information through natural conversation. Better yet, they cost a fraction of what a human support team would, making great customer service accessible to any business.
However, the fact that AI agents are cheaper than humans has led to many providers charging far more than the actual costs of running an agent. We think it's worth taking a closer look at the numbers.
How AI Support Agents Actually Work
Every AI agent runs on a Large Language Model (LLM) from one of the big providers: OpenAI, Anthropic, or Google. You've probably used their consumer products already (ChatGPT, Claude, Gemini). But out of the box, these models don't know anything about your business. To get responses tailored to your products and services, providers use a technique called Retrieval Augmented Generation (RAG).
RAG works by searching through the information you've provided (your knowledge base), retrieving what's relevant to a customer's question, and passing it to the LLM along with the question. The LLM then responds based on your actual content.
Every AI agent provider is doing this. They implement RAG, call an LLM, and package it in a nice interface. This is exactly what they should be doing, and it's what we do at Squidcom, because it works.
The Real Costs
Despite what many providers will have you believe, providing a great AI agent in 2026 is not rocket science. Get an LLM with RAG and you're 95% of the way there. And the cost of running an agent is really low, because it's almost entirely LLM calls. We're talking fractions of a penny per request. Which raises an obvious question...
Why Are They Charging So Much?
Intercom Fin charges $0.99 per resolution (successful conversation). Let's give them the benefit of the doubt and say they're using a high-quality model like Claude Sonnet 4. LLMs charge per token (roughly 1 token per word). According to Anthropic's pricing, Claude Sonnet 4 costs $3.00 per million input tokens and $15.00 per million output tokens.
Let's say a typical customer conversation involves a few questions and responses: about 100 words (130 tokens) from the customer, and 500 words (650 tokens) from the AI. The maths:
- Input cost: 130 tokens × $0.000003 = $0.00039
- Output cost: 650 tokens × $0.000015 = $0.00975
- Total: approximately $0.01
So when Intercom charges $0.99 for a conversation that costs roughly $0.01 to run, that's a 99x markup. Of course they have other overheads (infrastructure, support, development) but even accounting for all of that, the margins are hard to justify. And by the way, to access their AI agent you already have to be paying them flat monthly fees for their wider CRM system.
Intercom isn't alone in this. Many providers will try to lure you in with low starting costs, which scale poorly as your usage grows.
How We Do It Differently
Here at Squidcom, instead of saying, "Well, a human would cost you $1.20 per conversation, so we can get away with charging $1," we say, "Well, the LLM calls cost us $0.01 per conversation, so we'll charge $0.02."
We believe pricing should reflect actual costs, not what the market will bear.
Don't Just Take Our Word For It
At Squidcom, you can upload your documents and have an AI support agent up and running in minutes. It will answer questions specific to your business, with no complex setup or technical expertise required.
See for yourself how simple it can be.
Ready to see the difference? Try Squidcom free today. 