SafeSport-compliant messaging: built in, not bolted on

Every coach who works with kids has felt the tension, even if they have never said it out loud. Modern team messaging makes it dead easy to fire off a quick direct message to an athlete. Protecting those same athletes means that ease is the exact thing you have to handle with care. Easy and safe pull in opposite directions, and under time pressure, easy usually wins. That is the whole problem in one sentence.

Here is what it looks like in real life. A wrestler texts his coach at night, alone, one on one, because the app made a private thread the path of least resistance. Nothing bad has to happen for that to be a problem. The point of athlete-protection rules is that private, unseen, one-to-one contact between an adult and a minor is the environment you keep from forming in the first place. Not because every coach is a threat. Because the ones who are, count on exactly that environment, and a system that makes it the default has already lost the plot. A tool built for kids should make the open, visible path the easy one, and the private, unseen one hard to stumble into.

Retire the group-chat tools

Remind and Band can send a message. That is about all either was built to do, and neither was built for wrestling or for the responsibility that comes with a roster full of minors. You can retire both.

Scorecrypt keeps what they do, messaging your athletes and their parents, and adds two things they never had. The wrestling around it, so a message lives next to the roster, the schedule, and the event it is actually about instead of floating in a separate app. And the protection underneath it, so the safe way to communicate is built into how the tool works, not left to each coach to remember on their own.

Cutting one more tool out of your stack is the easy pitch. The one that matters more is the second one.

Built in, not bolted on

SafeSport compliant messaging to athletes and parents is on my build list, and I mean built in from the start, not a compliance mode I staple on after launch.

Protection designed in behaves differently than protection added later. When you build it in from the first line, the safe path is the default path. Nobody has to opt into doing the right thing, remember a setting, or flip into a special mode at 9pm when they are tired and just want to tell the team practice moved. When you bolt it on afterward, the safe path is always the extra step, and the extra step is the one that gets skipped. Every gap in these systems lives in that gap between what the tool makes easy and what it makes safe.

You can see the difference in how the two get built. Bolt-on compliance is a checkbox added after launch because a customer asked, sitting off to the side of a product that was designed to do the opposite. It works right up until the busy moment it was supposed to cover, and busy moments are the only ones that matter here. Built-in means the question "is this safe for a room full of minors" was in the first design conversation, not the fortieth. It shapes what the easy button does, which threads can even exist, and who can see what. You do not get that by patching. You get it by deciding it is the point before you write the thing.

This is also why I am saying it out loud this early, before every piece of it ships. Putting the principle in public is how you get held to it. It is a lot harder to quietly cut a corner on something you wrote a blog post promising.

I am still designing the specifics, and I am not going to pretend otherwise or put a checklist in your face that I have not shipped. When I make those calls, I will show my work right here, out in the open, the same as everything else. I am building to the athlete-protection standards that govern how adults communicate with minors in this sport, and I am not going to hand-wave them. When I lock the exact references, I will cite them right here. Getting that part exactly right is the job, not a footnote to it.

Where AI comes in, and where it does not

Now the question every athletic director is going to ask, so I am going to answer it before you have to. Where does AI come in, and is it reading my kids' messages?

Here is my line, and it is not moving.

AI in Scorecrypt messaging does one thing. It watches for signs of harm. Grooming patterns, talk of self-harm, the warning signs a program has a duty to see and act on. That is the entire job I am giving it.

It is not there to write your messages for you. It is not there to read conversations so it can sell you something, profile your athletes, or feed a model. It is pointed at one thing, protecting kids, and it is kept away from everything else on purpose.

I think that is the only defensible way to put AI anywhere near a tool full of minors. Not because AI is scary, and not because I am chasing "AI-powered" on a feature list. Coaches and athletic directors do not want an AI-powered anything near their athletes, and they are right to be skeptical. So I am not going to sell you the letters A and I. I am going to tell you the one job it does that earns its place, and name out loud all the jobs I am refusing to give it. Where I keep AI out is as much a part of the design as where I let it in.

And the one job I am giving it is not a gimmick, it is a duty. A coach is often a mandatory reporter, and even where the law is quieter, the responsibility is not. The hard truth is that the warning signs are usually there in the messages before anyone acts on them, and they get missed because no human is reading every thread on a roster of sixty kids, and no human should have to. A quiet check that surfaces a real concern to the right adult is the rare place where a machine genuinely helps protect a kid. That is why it earns its seat, and why it does not get to do anything else.

What to demand from any tool that touches your kids

I am building in the open partly so you can hold me to this. So here is a fair test, and I mean apply it to everyone, me included. If a vendor cannot answer these plainly, that is your answer.

Is the safe way to communicate the default way, or a setting someone has to find and turn on? Defaults are the whole game. Anything that depends on a busy coach remembering to be careful will fail on the night they are not.

Does the tool put AI, or anyone, in your athletes' conversations for any reason other than protecting them? If the answer wanders into personalization, engagement, or "improving the product," walk.

When you ask what standard they build messaging to, do you get a straight citation or a shrug? A shrug means they have not done the work.

Will they tell you what they have not built yet? A vendor who claims everything is finished and perfect is either lying or has not shipped to a real program. I would rather tell you the truth and let you decide.

Those questions are not a trap I set so Scorecrypt passes and everyone else fails. They are the questions I ask myself while I build this, and the reason I am comfortable putting them in front of you.

For the athletic director signing off on this

If you are an AD handing this to your coaches, you are the one who has to answer for how they communicate with kids. You should know exactly where the line sits before you trust any tool with that. So here it is, plainly. Communication designed to be safe by default. AI aimed only at catching harm, and deliberately blind to everything else. A standard I am building to on the record, not a marketing word.

I am not going to tell you this replaces your organization's own SafeSport training or your own policy. It does not, and any vendor who tells you their software does should worry you. What I will tell you is that the software should make your policy easier to follow, not harder. That is the bar I am building to, and I am building it where you can watch.

One more tool off your stack, one more thing you do not have to worry about at 9pm, and your athletes better protected for it.

See what I'm building. Start your free trial. No credit card.