An AI practitioner on why human connection matters more - not less - in an agentic world.
I'm Not Writing This Because I Just Discovered AI
Let me be clear about where I'm coming from.
I run opencream.ai, where we help European B2B companies - especially German Mittelstand "hidden champions" - integrate AI into their operations. I'm a fractional Chief Commercial Officer for biotech companies that have been using AI for molecule design, drug discovery, and protein folding for years. I've been using Claude and ChatGPT daily for over two years.
AI isn't something I read about. It's something I work with. Something I help companies adopt. Something I've watched transform from "interesting experiment" to "I don't know how I worked without this."
So when Matt Shumer published his essay "Something Big Is Happening" - his unfiltered take on where AI is heading - I didn't read it as news. I read it as confirmation of what I've been seeing from the inside.
But that's not why I'm writing this.
I'm writing this because working with AI every day has changed how I think about raising my kids.
The Breakfast Table
Last Tuesday, my 14-year-old asked me to help with his English homework while scrolling TikTok. My 12-year-old was watching rugby highlights on his phone. My 10-year-old daughter was negotiating more iPad time.
Normal morning. Three kids, three screens, one dad trying to get everyone to school on time.
My kids speak German, French, and Spanish. They're learning English at school. Four languages before any of them have finished secondary school. That's not nothing.
But here's what I've realized from working with AI: the languages aren't valuable because of translation. AI handles that now. The languages are valuable because of what they represent - the ability to connect across cultures, to signal that you cared enough to learn, to build trust in ways no algorithm can replicate.
And that insight - that human connection becomes MORE valuable in an agentic world, not less - is shaping how I'm raising them.
What Tristan Harris Taught Me
Before I go further: my thinking about this didn't start with Matt Shumer's essay. It started with Tristan Harris.
Years ago, Harris - the former Google design ethicist - started warning about the attention economy. How our phones are designed to capture focus. How social media optimizes for engagement, not wellbeing.
I watched his talks. I read his work. And I started paying closer attention to what was competing for my kids' attention - and what I could do about it.
The TikTok scroll isn't just entertainment. It's a product designed by teams of engineers to capture attention. The algorithm knows what makes a 14-year-old keep watching better than the 14-year-old does. That's not scary - it's just the landscape we're navigating.
Now add AI that can generate infinite personalized content. AI companions that are always available, infinitely patient, optimized to engage. AI tutors that never get frustrated. AI friends that never judge.
This isn't doom. This is just the world. And understanding it changes how I parent.
Harris helped me see that I can't control the technology landscape - but I can teach my kids to navigate it. I can help them understand the difference between being a user and being used. I can model healthy relationships with these tools.
The Gap Between Silicon Valley and Suburban France
I live in France. I work with European companies. And I can tell you: the gap between what's happening in San Francisco and what's happening here is enormous - and getting wider.
Matt describes telling AI to build an app, walking away for four hours, and coming back to find it done. Done well. Better than he would have done it himself. He's not exaggerating. I've experienced versions of this in my own work. The latest models don't just execute instructions - they make decisions that feel like judgment.
But here in Europe, most companies are still debating whether to "explore AI options."
The mindset difference is stark:
San Francisco: "How fast can we deploy this end-to-end?"
Europe: "Should we form a committee to evaluate this?"
I've sat in boardrooms with biotech executives who think they're ahead because they use Copilot to summarize their meetings. That's like thinking you're ahead in the smartphone race because you have a flip phone with a color screen.
The irony is that European companies - especially the German Mittelstand, the hidden champions, the specialized B2B suppliers - have exactly the domain knowledge that could make AI integration incredibly valuable. They have 40 years of expertise sitting in people's heads. But that expertise is 5-10 years from retirement. And the window to capture it, to multiply it with AI rather than lose it forever, is closing.
What School Teaches vs. What I'm Adding
My oldest (14) is studying for exams that test his ability to memorize and reproduce information. He's learning to write essays in a specific format. That's fine - it builds discipline and structure.
But AI already writes better essays than most students. So at home, we talk about what essays are actually FOR: learning to think clearly, to structure an argument, to communicate. The output matters less than the process.
My son (12) is figuring out what career he might want. The guidance he gets points toward traditional paths: law, medicine, engineering, finance.
These are exactly the categories Matt Shumer identifies as most exposed in the coming years. So at home, we talk less about job titles and more about what he's curious about. What problems interest him. The specific roles will change - the underlying interests won't.
My daughter (10) is learning the basics. Reading, writing, arithmetic. The fundamentals that supposedly prepare you for everything else.
The fundamentals still matter. But what matters more is learning HOW to learn - because she'll be learning new things her entire life, in ways we can't predict.
The Education Gap That Already Existed
This isn't the first time schools have failed to prepare kids for reality.
My 14-year-old can solve quadratic equations. He has no idea what compound interest is.
My 12-year-old knows the dates of historical battles. He doesn't know the difference between an asset and a liability.
My 10-year-old is learning fractions. Nobody is teaching her that money is a tool, not just something you spend.
This gap has existed for decades. Schools teach kids to be employees - show up, follow instructions, get a grade. They don't teach kids to think about money as something that can work for you. They don't explain how credit can be a tool for building wealth, not just a trap for buying things you can't afford. They don't cover the basics that separate people who build security from people who live paycheck to paycheck.
I've had conversations with my kids about compound interest that felt like I was explaining magic. "Wait, the money makes more money? Without doing anything?" Yes. That's the point. That's the entire game. And you're 14 and this is the first time anyone's told you.
The financial literacy gap was already a scandal. Now add AI to the list.
Schools aren't teaching kids about money. They aren't teaching them about attention and how their minds are being monetized. And they aren't teaching them about AI - or preparing them for a world where the career paths they're aiming for might not exist by the time they graduate.
It's not one gap. It's a stack of them. And each one makes the others worse.
A kid who doesn't understand compound interest won't understand why starting to invest at 22 instead of 32 matters so much. A kid who doesn't understand the attention economy won't see how TikTok is the product and they're the resource. A kid who doesn't understand AI won't recognize that memorizing information is training for a race they've already lost.
Schools are preparing kids for a world that existed 20 years ago. The money part was already outdated. The AI part is making it obsolete.
Teaching Them About Attention
My kids live in a world designed to capture their attention. TikTok, Snapchat, YouTube, Xbox - every platform has teams of engineers optimizing for engagement. I know this. I work in tech.
So we talk about it openly. Not "screens are bad" - that's not true and they know it. More like: "This app is designed to keep you watching. Do you feel like you're choosing to watch, or like you can't stop?"
I still catch my 14-year-old scrolling at 11pm when he should be sleeping. That's normal. But he also understands WHY it's hard to stop. He's not just fighting willpower - he's learning to recognize when something is designed to override his choices.
That awareness is the skill. Not perfect self-control, but understanding the game you're playing.
Why I Push Languages (Even When They Push Back)
My kids speak German, French, and Spanish. They're learning English at school. That's four languages - and sometimes they wonder why it matters.
"Dad, my phone can translate anything. Why do I need to speak German with you?"
It's a fair question. AI translation is remarkably good now. Apple has AirPods that translate in real-time. You can technically "communicate" with anyone in any language.
But communication isn't just understanding words. It's connection through words.
When I speak German with a client in Munich, something happens that doesn't happen when I speak English. There's a shared cultural context, a mutual vulnerability of speaking someone's language, a signal that I cared enough to learn it. The relationship is different.
Same with my broken Spanish. I can run a business meeting in Spanish - badly, with mistakes, searching for words. But the fact that I'm trying, that I'm willing to be imperfect in their language rather than demand perfection in mine? That builds trust faster than any translation app.
My kids live in France. They naturally default to French. Speaking German with me or taking the extra language option at school feels like unnecessary effort. I get it. But I keep pushing because I understand something they don't yet: in an agentic world where AI handles information tasks, the human connection skills become the differentiator.
I've watched my daughter explain something to a friend using only hand gestures and laughter. No words at all. Perfect communication.
AI can translate. It can summarize. It can even mimic emotional tone. But it can't replicate the moment when two people really see each other across a cultural gap and find connection anyway.
That's what languages teach - not vocabulary, but the skill of reaching across difference.
What I See From the Inside
As a fractional CCO, I'm on the commercial side - I don't design molecules or run the AI models. But I work inside biotech companies that have used AI for molecule design, drug discovery, and protein folding for years. AlphaFold changed structural biology. AI-driven drug discovery has produced candidates now in clinical trials.
That's "headline AI" - the breakthrough stuff that makes the news.
What's different now is operational AI. The day-to-day stuff. And that's where opencream.ai comes in - helping European B2B companies, especially the German Mittelstand hidden champions, integrate AI into their actual operations.
I know companies where the sales team has 30 years of customer knowledge in their heads - every client's quirks, every relationship's history - and none of it is documented. I know operations managers who can diagnose a problem by the sound a machine makes, and that expertise will retire with them.
We help them capture that knowledge, multiply it with AI, prepare for what's coming. The hidden champions I work with are actually more open to this than you'd expect - certainly more than the chemical industry where I started my career. They see the value.
And here's what my professional work has taught me about parenting: the things that are hardest to capture in AI - relationships, judgment, context built over years - those are exactly the skills that become more valuable.
The window for companies to capture institutional knowledge is closing. But the window to raise kids who thrive in an agentic world? That's still wide open.
What I'm Actually Doing
I don't have all the answers. But working with AI every day has shaped how I parent:
Curiosity over career paths. I've stopped asking "what do you want to be?" and started asking "what are you curious about?" The job titles will change. The interests underneath won't.
Languages for connection, not translation. My kids already speak four languages. The challenge is keeping them motivated when they live in France and could just default to French. I keep pushing because I've seen what happens in business when you speak someone's language - even badly. Trust builds differently.
Honest conversations about attention. Not "screens are bad" - they know that's not true. More like: "This app has teams of engineers trying to keep you watching. How does it feel when you can't stop?" Awareness is the skill, not perfect discipline.
Modeling adaptation. My kids watch me use Claude and ChatGPT daily. They see me experimenting, learning, getting confused and figuring it out. That's more valuable than any lecture about "preparing for the future."
Active vs. passive screen time. Using AI to build something is completely different from scrolling TikTok. One prepares them for an agentic world, one makes them a product for advertisers. We talk about the difference.
Financial basics that school skips. Compound interest, assets vs. liabilities, how credit works. The same gaps existed 20 years ago. I'm not waiting for schools to catch up.
And in my work, I spend an hour every day pushing AI into new territory. The things it kind of does today, it will do well in six months. Staying current isn't optional - it informs everything I do, including parenting.
The Real Opportunity
Matt Shumer ended his essay by saying the future has already arrived - it just hasn't knocked on most people's doors yet.
For my kids at that breakfast table, the future is already in their pockets. And because I work with AI every day, I can help them see it clearly.
They might think TikTok is just entertainment. I can show them how it's designed to capture attention - and that understanding gives them power.
They might think school is preparing them for careers. I can help them see that curiosity and adaptability matter more than any specific job title.
They might think their four languages are unnecessary because phones can translate. I can explain why the connection across cultures is the point, not the words.
I'm not writing this because I'm worried. I'm writing it because I'm optimistic - about AI, and about raising kids who can thrive alongside it.
The schools will catch up eventually. The governments will figure it out. But we don't have to wait. Parents who understand AI can start now, at breakfast tables, with honest conversations.
Something big is happening. And our children can be ready for it.
That's the opportunity.