Kidas’ ProtectMe software listens to your kids

Here’s a question for you. How do you ensure that you are raising your child’s internet time without violating their privacy? kidas may have an answer for you, with its ProtectMe software. ProtectMe isn’t just a piece of babysitting software that alerts parents when their own child is misbehaving. It also recognizes signs of incoming harassment.

The program uses artificial intelligence to track alarming behavior between children in more than 120 games. ProtectMe specifically searches for context; it can tell the difference between a playful jab and an actual threat.

At the end of each week, the software generates a report for the parents. This report warns them of any worrisome behavior by – or towards – their child.

It sounds good, in theory. But I had a few questions about the details. So I had to ask Kidas boss Ron Kerbs a few questions. Just a few questions about how the software works exactly and where it goes in the future.

How it currently works

GamesBeat: Is ProtectMe limited to voice chat only?

Ron Curbs: The ProtectMe software provides both voice and text monitoring. It knows when to turn on and off based on when games or communication apps like Discord are turned on and off.

GamesBeat: Is it only for users? That is, the main purpose of ProtectMe is to observe the child’s outgoing speech and not the incoming speech?

curbs: ProtectMe has the ability to understand communication both inside and out, which means we know if a child has been exposed to, for example, cyberbullying, as a victim, perpetrator or witness. For privacy and compliance reasons, we do not provide gamertags, personal information, or transcripts to parents or guardians. Only in the harshest circumstances will we provide additional context, for example if there is a clear potential for trafficking or if the child intended to meet a stranger in person, among other critical circumstances.

GamesBeat: Can the software handle audio splitting? If a microphone is connected through an audio mixer or through a digital audio cable, can ProtectMe detect that the default Windows option is not in use and adjust it accordingly?

curbs: Kidas supports splitting audio; in some cases we analyze more than eight channels simultaneously. Our technology can distinguish between relevant data and background noise that we can ignore.

Context matters

GamesBeat: Can ProtectMe tell the difference between a child actively saying something to someone and reading something aloud? Between another user talking and a recorded video?

curbs: The new thing about our technology is that it understands the context as part of the game and goes beyond language processing and keyword recognition. We all know that trashtalk is part of the gaming experience; however, the difference with ProtectMe is that because our AI software has been tested with thousands of scenarios, it can respond correctly every time and decipher whether an interaction represents a direct threat or benign nonsense.

That’s pretty incredible and something that no one else on the market offers. We also know that, in general, what is appropriate to expose, say, a 7-year-old to is very different from that of a 14-year-old, so with the guidance of child psychologists and cyberbullying experts, we offer a different threat severity scores and recommendations. based on age. When someone uses a voice changer or recording, we listen to the context and words being spoken, not the depth or pitch of the words spoken.

Privacy remains important

GamesBeat: Does ProtectMe recognize repeated voices? If a child routinely interacts with someone who is regularly identified as a bully, does Kidas warn the parents their child is interacting with?

curbs: No, we are incredibly vigilant about privacy laws and security. We encourage parents who see repeated threats to continue talking within their families. Since these topics can be incredibly challenging for parents to discuss, we offer personalized recommendations for how to talk to their children about what happened.

GamesBeat: Does ProtectMe track and/or provide updates for parents over time? For example, if ProtectMe flags a child 5 times a week and reports it to the parents, and the following week ProtectMe flags a child 3 times, is the parent alerted to the better behavior?

curbs: We keep track of the changes from week to week, but not quite like that. Our reports start with an overall color scoring system to categorize the severity of threats detected in that week:

Red: immediate action required; The most disturbing threats like contact with a pedophile.
Orange: action required; For related threats such as bullying.

Yellow: worth noting; Behaviors you should be aware of, such as age-appropriate trash.

Green: nothing wrong: no indication of relevant threats found.

Our reports include recommendations based on the overall severity score and the threat detected. When we are repeatedly exposed to the same threats, we provide a set of unique recommendations based on detected historical threats.

From there, we also provide the time spent playing analytics. These show which games were played the most, how many hours were played each day and in total for the week, how this compares to last week, and how the total time spent playing compares to all ProtectMe gamers. We are advocates of screen time and want to keep gaming safe and fun; however, parents have told us that they never know how much their kids actually play and how it compares to other gamers. They now have even more insight than without the software.

What the future looks like

GamesBeat: Will ProtectMe ever be able to monitor game-specific harassment? (e.g. a user spamming pre-setat features in Rocket League such as “what a save!”; again and again to mock a player)

curbs: This is not a current offering, but we have plans to expand our analysis to include game-specific harassment. For example, as part of our partnership with Overwolf, we use Overwolf’s APIs to access game-specific events to detect those situations.

GamesBeat: Will there be more in-depth tracking for ProtectMe in the future? (e.g. tea bags, counter-attack sprays, etc.)

curbs: While our current focus is on audio and text chats, detecting other toxic events that don’t involve peer-to-peer communication, such as tea bags and smurfs, are definitely on our roadmap.

We also partner with some of the largest game studios and gaming equipment manufacturers to provide them with custom solutions that fit their unique technology, storylines and challenges.