Media outlets love a "manifesto." They crave the dramatic digital trail—the timestamped emails, the frantic texts to relatives, the sudden burst of anti-government vitriol sent minutes before a tragedy. It provides a comfortable, linear narrative. It suggests that if we just monitored the right inbox or flagged the right sequence of words, we could stop the clock.
This is a dangerous fantasy.
The obsession with "warning signs" sent moments before an attack is a form of survivorship bias that cripples actual intelligence work. We focus on the communication that occurred because it is visible, not because it is the most significant causal factor. By the time a gunman hits "send" on a final message to his family, the failure is already total. That message isn't a lead; it’s an obituary for a prevention strategy that never existed.
The Signal-to-Noise Nightmare
Most analysts operate under the "volcano model." They assume pressure builds visibly until an eruption occurs. They want to believe that radicalization follows a tidy, measurable path that mirrors a marketing funnel.
It doesn't.
In the real world, we are drowning in "anti-government sentiment." If you used AI to flag every individual in the United States who sent a vitriolic, anti-state message to a family member this morning, you wouldn't have a list of suspects. You would have a database of forty million people.
The industry consensus is that we need more data, more scraping, and more "predictive" linguistics. I’ve watched agencies sink nine-figure sums into sentiment analysis tools that couldn't distinguish between a frustrated taxpayer and a genuine threat if their funding depended on it. The "lazy consensus" ignores the fact that intent is not a linguistic trait. It is a behavioral one.
The Fallacy of the Final Message
When the press highlights that a shooter sent a message "minutes before" an event, they imply a window of opportunity. This is intellectually dishonest.
A message sent five minutes before an opening shot is not actionable intelligence. It is noise. Unless a law enforcement officer is already standing in that person’s living room, that data point is useless for prevention. Yet, the post-game analysis always centers on these digital crumbs. Why? Because it’s easier than admitting that our current behavioral models are fundamentally broken.
We are looking for the "what" (the text) instead of the "how" (the logistics).
The Distinction Between Radicalization and Mobilization
The biggest mistake in current security discourse is conflating ideology with action.
- Radicalization is a change in belief. Millions of people hold radical, even hateful, views without ever lifting a finger.
- Mobilization is the transition to tactical preparation.
Focusing on "anti-govt writings" is a radicalization metric. It tells you what someone thinks. It tells you nothing about their capability or their timeline. If we want to disrupt these cycles, we have to stop patrolling the world of ideas and start looking at the world of procurement and reconnaissance.
The person who buys high-capacity magazines and spends six hours staring at a facility's entrance on Google Maps is a threat. The person venting to their uncle about the "Deep State" is just another Saturday morning on the internet. By prioritizing the writings, we dilute our resources and miss the movement.
Why AI Won't Save Us
The tech sector promises that LLMs and "pattern recognition" will bridge this gap. They are lying to you to sell subscriptions.
Algorithms are great at finding "more of the same." They are terrible at identifying the "black swan" event—the individual who doesn't fit the previous profile. Every time a shooter sends a last-minute email, the tech giants tweak their filters to catch those specific keywords.
This is the equivalent of trying to win a race by looking exclusively in the rearview mirror.
The "contrarian truth" is that more surveillance often leads to less safety. When you increase the volume of data points, you increase the probability of false positives. You create a "Cry Wolf" environment where genuine signals are buried under a mountain of digital angst.
The Cost of the Narrative
When we fixate on the "anti-govt" angle, we also grant these individuals the one thing they want: a legacy.
By dissecting their "writings" as if they are meaningful political treatises, the media validates the shooter's self-image as a revolutionary. We treat a psychopathic act as a logical conclusion to a political grievance.
This creates a feedback loop. Potential imitators see that their internal misery can be rebranded as a "cause" if they just leave the right digital trail behind. We are effectively providing the template for the next "manifesto."
Stop Looking for "Why" and Look for "When"
If you want to actually disrupt the trajectory of violence, you have to ignore the rhetoric.
I’ve seen security teams ignore massive red flags—unexplained large purchases, sudden tactical training, the abandonment of long-term assets—simply because the individual didn't "sound" like a radical. Conversely, I've seen lives ruined because someone’s edgy Discord logs were flagged by an overzealous algorithm that couldn't recognize irony.
We need to pivot.
- Ditch the Sentiment Analysis: Stop trying to "solve" the person's brain. You can't.
- Hard-Target Logistics: Focus on the friction points of an attack—weapon acquisition, site casing, and dry runs.
- De-Platform the Grievance: Stop publishing the contents of the "last-minute emails."
The industry is currently built on the premise that we can predict the human heart by reading its digital exhaust. It is a lucrative lie that keeps us reactive, vulnerable, and perpetually surprised.
The message sent minutes before a shooting isn't a "missed clue." It’s the sound of the system crashing. If you're waiting for the email, you've already lost.
