Silicon Valley's Violence Problem Isn't What You Think

---

A 20-year-old from Texas threw a homemade bomb at Sam Altman’s gate. That’s the headline everyone’s reading. But the real story is scarier: he had a written manifesto about AI threats, a list of other tech executives, and federal felony charges stacked on top of attempted murder.

This isn’t some random lunatic. This is what happens when apocalyptic AI discourse meets real people with grievances and access to explosives.

Photo by Polina ⠀ / Pexels

The Rhetoric-to-Reality Pipeline

Let’s be clear about what’s happened. Over the past three years, AI safety discourse—which started as a reasonable philosophical debate about existential risk—has metastasized into something that sounds, to the uninitiated, like a call to action.

When researchers, journalists, and yes, some AI executives themselves, spend thousands of hours discussing scenarios where artificial intelligence poses an “existential threat” to humanity, they’re doing something important. Thinking ahead matters. But when that reaches the general population, it doesn’t arrive as nuanced academic argument. It arrives as: these people are building something that could kill everyone, and they’re not stopping.

The Texas attack is what you get when someone takes that messaging seriously and decides Sam Altman is ground zero.

This hasn’t happened in a vacuum. We’ve seen similar pattern-matching before. The Unabomber justified his violence through a manifesto about technology’s corrosive effects. QAnon believers stormed the Capitol. The pipeline from online radicalization to physical violence has been proven repeatedly. We just haven’t seen it weaponized against tech executives until now—and apparently, it’s happening twice. Rockstar Games got hacked again, this time by English-speaking attackers who had clearly done their homework.

My read: the industry treated its critics as outside agitators for too long. Now some of those critics are treating it as an existential threat worth stopping by force.

Photo by nappy / Pexels

Why Amazon’s $10.8B Bet Is the Real Power Play

Here’s what I find more interesting than the violence: Amazon just spent $10.8 billion buying Globalstar to expand its satellite internet service. That’s not a technology investment. That’s a geopolitical move.

Elon Musk built Starlink as a private space network that happens to provide internet. Amazon’s taking the inverse approach: they’re buying existing infrastructure and folding it into a giant e-commerce and logistics empire. The endgame isn’t selling internet to rural Montana. It’s controlling the last-mile delivery network that doesn’t depend on terrestrial infrastructure—weather, terrain, or government permission.

Compare this to 2008, when Amazon Web Services started as a side project and became worth more than the entire retail business. Satellite internet will follow the same pattern. What starts as “reliable connectivity” becomes the backbone for everything else—delivery drones, autonomous logistics, real-time data services.

Musk’s sitting on Starlink, trying to take SpaceX public. Bezos just cut the line and bought his way to parity. That’s not innovation; that’s Amazon being Amazon.

The Quantum Europe Question Nobody’s Asking

Europe’s supposedly in the race for quantum computing leadership. The headline asks whether it could win. The answer is no.

Not because Europe doesn’t have talent—it does. But because quantum computing isn’t actually a technology race yet. It’s a capital race, and America and China are running on a different track. The U.S. has Google, IBM, and a dozen VC-backed startups burning through billions. China has state backing and a 10-year plan. Europe has… promising companies?

Quantum computing will be genuinely transformative whenever it actually works at meaningful scale. We’re probably 5-10 years away from that. By then, whoever controls the hardware controls the protocols, the software stack, and everything downstream.

Europe’s best play isn’t winning the quantum race. It’s regulating it in ways that force American and Chinese companies to license their technology to European firms. But that would require actually being ruthless about antitrust, and frankly, Europe’s proven it doesn’t have the stomach for that.

Elon’s On TikTok Now, and That’s Hilarious

Elon Musk joined TikTok. Also Instagram. Also presumably he’s about to announce his presence on BeReal and Mastodon and whatever other apps help him feel like he’s at the center of attention.

The stated reason: he’s prepping SpaceX for a public offering. The real reason: he can’t help himself. The man owns X (formerly Twitter), fired most of its staff, watched the platform’s value crater, and his response is to start ghost accounts on competitors’ platforms.

This is what billionaire desperation looks like. It’s not dangerous or even particularly interesting. It’s just… sad in a way that’s oddly compelling. He’s checking if other platforms will let him be as visible, as unfiltered, as he was when he had actual power.

They will. And it won’t matter.

Roblox’s Age-Check Trap

Roblox is extending age verification to create two-tiered accounts. With 144 million daily users—mostly kids—this is important. Also impossible.

Age verification on the internet has never worked. Not because the technology is hard, but because people lie, use their parents’ IDs, and generally don’t care about terms of service designed “for their own good.” Roblox knows this. Parents know this. So why announce it?

Because Roblox is getting older. It needs adult legitimacy. This announcement isn’t really for kids or parents. It’s for regulators and investors who want to see the company taking safety seriously. Whether it actually works is secondary to whether Roblox can point at it and say: “We tried.”

I think this is fine. I also think it’ll be circumvented within months and we’ll never hear about it again.

The Actual Story

The violence against Altman matters because it’s a pressure release valve. Someone took the existential-risk-to-humanity messaging seriously enough to act. That’s not actually that surprising—we’ve known for years that online radicalization has real-world consequences.

What’s surprising is that tech leadership hasn’t course-corrected. OpenAI is still gating access to GPT-5.4-Cyber, ostensibly for safety. The safety-through-restriction model has become industry gospel. But the attack on Altman suggests that transparency and engagement might actually reduce violence more than walls do.

Meanwhile, the real power is consolidating. Amazon’s playing 4D chess with satellite infrastructure. Musk’s publicly admitting his platforms are dying by jumping to competitors. China and America are building actual quantum capabilities while Europe comments on the sidelines.

The bomb at Altman’s gate is the least important story here. It’s a symptom, not a cause.

What I’m Watching

• Copycat attacks through Q3 2025: If we see even one more direct action against AI executives, the whole industry’s security posture changes overnight. Watch for whether companies publicly acknowledge threats or stay silent.

• Amazon’s satellite deployment timeline: Bezos said SpaceX public offering prep was 2025. If Amazon’s Globalstar integration moves faster than Starlink’s public debut, that’s the signal that corporate capital wins this arms race, not founder obsession.

• OpenAI’s GPT-5.4-Cyber adoption rate: They’re positioning this as a “trusted partners only” release like Anthropic does. If it spreads beyond 50 companies by fall, the whole “limited release” model was theater. If it stays locked, OpenAI’s admitting they’re afraid of what happens when this tool gets loose.