Guardrails & protections
Most AI assistants either say no to everything, which makes them useless, or yes to everything, which makes them dangerous. Fred handles the risky edges automatically, before a word reaches your visitor. That is exactly what frees it to be fast, warm, and genuinely helpful on everything else.
Think of guardrails the way a race car driver thinks about brakes. Better brakes do not slow you down, they let you go faster, because you trust the car to stop when it has to. Fred is the same. Because the rare, high-stakes moments are caught and handled the instant they happen, Fred can move quickly and sound human on the thousands of ordinary questions that actually grow your business. You get more booked jobs and fewer messes to clean up, at the same time.
Every Fred is protected by four layers that build on each other. You do not have to assemble them. The right ones switch on for your business and your country automatically.
Always on, everywhere, for every customer. Someone in crisis, a threat of violence, anything involving a child, fraud, hate, or a request for sensitive information like a Social Security or card number. These are handled the same way no matter your industry or country.
A crisis hotline in London is no use to someone in Texas, and a privacy law in Europe is not the same as one in California. Fred loads the right emergency numbers and the right privacy rules for where you operate. See the regional rules →
A dentist, a roofer, and a law firm each have things they must never say. Fred runs a pre-built pack tuned to your field, so it knows where your lane ends before anyone tests it. See it by industry →
The pack is the floor, not the ceiling. You can add unlimited rules of your own in plain language, block, warn, or quietly log on any phrase you choose. No developer required.
Every layer runs on our servers, before Fred replies. It is not a polite suggestion buried in a prompt that a clever question can talk around.
These protections ship with every deployment and cannot be switched off. They exist because the cost of getting one of them wrong, even once, is far bigger than any single sale.
Someone in crisis
If a person hints at suicide or self-harm, Fred stops selling, shares real crisis help lines right away, and brings in a human. It never tries to play counselor.
A threat of violence
Talk of harming people or making a weapon is refused on the spot and flagged for your team. Fred will not engage with it, and it will not coach it.
Anything involving a child
If a conversation touches the safety of a minor, Fred hands it straight to a human. No automated reply is the right reply here.
Fraud, hacking, or anything illegal
Requests to commit fraud, break into accounts, or anything unlawful are declined, with an offer to connect the visitor to your team for lawful options.
Hate, harassment, or explicit content
Fred keeps the conversation respectful and on-topic. It will not produce hateful, harassing, or sexual content, no matter how it is asked.
Sensitive personal information
Fred will not collect a Social Security number, card number, or bank details in chat. It points the visitor to a secure channel instead, so that data never lands in a transcript.
Always honest about being AI
Fred tells people they are talking with an AI assistant and offers a human for anything important. Honesty here is now the law in a growing number of places.
Three situations can quietly turn into a headline, a lawsuit, or worse. Here is why each one matters and exactly what Fred does about it.
People in real distress reach out to whoever is there, even a furnace-repair chat at 2am. The wrong response is unthinkable.
An assistant that engages with a threat, even to argue, is a liability. Fred treats these as hard stops.
Customers who are blind, deaf, or use assistive tech have a legal right to equal access, and inaccessible sites draw real lawsuits.
See how the crisis, accessibility, and stay-in-your-lane protections work →
A privacy law in Europe is stricter than one in the United States. A crisis line in Canada is a different number than one in the UK. Fred loads the right emergency resources and the right privacy protections for where you operate, so a visitor always gets help that actually works where they are.
Talk to Fred.
Go live in about 20 minutes. Or ask Fred a question right now, it is in the corner of this page.
