The OODA Network on New AI-based Threats and Issues for the U.S. Cognitive Infrastructure, Internet Safety and Wellness for Kids

Following the initial discussion at the April 2024 OODA Network Monthly Meeting of Domain-Specific Large Language Model Development and Use Cases, and the follow-up discussion on Exponential Technology Disruption, Global Risk, and Geopolitical Futures, the spoofing capabilities of AI creating new threats and issues began a discussion that gradually focused on and distilled more granular insights related to Internet-based Child Safety and Wellness  – and potential preventive business models and platforms. Details of that portion of the meeting discussion can be found here.  

Artificial Intelligence: New Threats and New Issues for the U.S. Cognitive Infrastructure

A Discussion of Internet-based Child Safety and Wellness  – and Potential Preventive Business Models and Platforms

• The meeting highlighted the challenges posed by artificial intelligence and the need to adapt to these rapidly evolving technologies  – comparing these challenges with past issues like online chatting safety, in early newsgroups and AOL chat rooms for example, and learning how to address these new threats effectively.
• The conversation turned to the need to understand and accelerate the right convergence to address emerging threats and bring advancements into enterprises, especially proactive measures to protect children and possible entrepreneurial startup models.   
• One network member brainstormed the potential for a startup to create a system that monitors children’s online activity and alerts both the child and the parent about concerning behavior.
  • This idea stems from the challenge of monitoring children constantly and the belief that AI connected to a broader engine could effectively identify threats such as stalkers or criminals.
  • The proposed value proposition of this AI-based monitoring business model was discussed as the ability to monitor children’s online activity using AI to detect risks more effectively than manual monitoring.
• A network member mentioned a company, Bark, as an example but called attention to the need for improvement in accuracy, citing an incident where the AI flagged a child’s innocent conversation about being homesick as concerning.
• Bark Case Study
  • What does Bark do?  Bark is a parental controls company that helps parents protect their children online and in real life – with  a suite of products that families can choose from:
    • Bark Phone
    • Bark Premium & Bark Jr (apps)
    • Bark Home
  • Content Monitoring
    • Bark monitors your child’s texts, email, YouTube, and 30+ apps and social media platforms for issues like cyberbullying, adult content, sexual predators, profanity, suicidal ideation, threats of violence, and more.
    • Parents receive text or email alerts only when something potentially problematic occurs online. You won’t have full access to everything on your child’s phone — just the things you might need to know about. This saves you countless hours of scrolling through every online interaction your child has while allowing you to build trust and maintain an open line of communication with your child.
  • Screen Time Management:  Bark’s screen time management features help families set time limits and create schedules for when their children’s devices can connect to the internet
    (through both cell service and Wi-Fi).
  • App and Website Blocking:  Our website blocking feature lets you select which apps and websites are appropriate for your child to access on their devices. You can block specific sites or even whole categories like streaming services, online gaming, sexual content, and more.
  • Location Sharing:  Bark’s location sharing features include both location alerts and check-ins. Through location tracking alerts, you can get notified when your child arrives at or leaves a specific place. With check-ins, you can see exactly where your kid is at a given time.
• A network member shared an “old man rant,” expressing the belief that there is not more evil or terribleness in the world today compared to the past.
  • The speaker emphasized that the amplification of negative events, even as small as a kid stubbing their toe, creates the perception of increased negativity.
  • Despite acknowledging the connectivity of good and bad through the internet, the speaker highlighted a sense of continuity in the human experience over time – that it has been impossible to and there has been no need to monitor children 24/7 over the millennia. 
  • He shared a common experience for many people on the call that grew up in the U.S. in a bygone era – that of time after school or for even longer summer days, with parents not even really knowing where we were for the day, really, and not needing to be home “until the streetlights came on before dinner time.”
• Another network member shared concern about AI managing the therapy business and the challenges within the field. 
• Another member expressed a preference for talking to AI over a friend’s psychotherapist due to perceived issues in the industry.
• A participant on the call offered a framing:  “This is similar to the rise of chatting on an OL in the early nineties. So, the idea that threat actors could now reach our kids and chat with them, potentially bring them into a threatening situation, it took time for us to learn how to deal with those threats both socially and broader as a country. We are doing the same thing now with this “spoofing of sorts with AI. So, for example, my oldest child is is turning four in July – and we have a code word.  If there is an “event,” –  I am instructing my kids at a very early age with a little bit of tradecraft, such that if there is a problem and I have a video call with them, I can identify whether or not that’s actually them.”  
• The membership discussed various ways our collective intelligence and subject matter expertise could address these new threats and issues, with the same network member above continuing, “This problem’s only going to persist as my children get older.  I’m not sure if what I am doing with my kids is necessarily going to be the solution for all parents. Still, part of an area where I think we can actually have real value outside of our respective networks is providing a little bit of guidance on these technologies.” 
  • The network agreed, with a formative position statement: “These technologies should continue to be developed, but here are some of the threats we are seeing come out of this. And here are potential solutions for you – if you are a board member, if you are a parent, if you are a member of the DoD, and you are trying to use these tools like just a little bit of OPSEC/OPSWAT. 
    • “We could at least “implant” these ideas using our collective voice.”
    • “And if a certain audience takes one thing away that they didn’t have before, that’s a win.”
• The conversation then turned to Jonathan Haight’s book “The Anxious Generation,” which includes an overarching theme that the overprotection of the physical world and underprotection of the virtual world lead to a rise in mental illness among teens and children.  This issue was described as causing significant distress, overwhelming some children and parents. 
  • The Anxious Generation: How the Great Rewiring of Childhood Is Causing an Epidemic of Mental Illness by Jonathan Haidt  – https://a.co/d/1kgNlVC 
  • A member reinforced the unchanged nature of the situation, indicating ongoing challenges and the need for increased awareness and proactive measures to address potential risks.  
  • In contrast, but not without empathy and compassion for the climate for children and parents being discussed, the lack of creativity and thinking amongst individuals, potentially making them targets without realizing it, was voiced as an overall behavioral concern of the individual over the affordances and phenomenology of the network and platforms.   
• From the Meeting Chat 
  • Tim Estes is building a company that aims to protect children in social media (and other “digital” places): https://angelkids.ai/
  • OODA Network Member Spencer Ante’s piece in Fast Company was mentioned – POV: Why pausing AI development is a bad idea
    https://www.fastcompany.com/90874897/pov-why-pausing-ai-development-is-a-bad-idea  – “There’s no doubt AI should be developed safely and transparently, but putting the most critical technology of our age in a timeout could weaken our country at a critical moment.”  
  • Is Google Making Us Stupid:  https://www.theatlantic.com/magazine/archive/2008/07/is-google-making-us-stupid/306868/
  • talk2us.ai: AI Chatbot for Mental Health, Relationship, Family, and Personal Development Counseling

The OODA Network on Exponential Technology Disruption, Global Risk, and Geopolitical Futures

<