AI growth sparks verified risks and unverified backlash claims

A 20-year-old Texas man was arrested earlier this year after allegedly attempting to set fire to OpenAI’s headquarters and the home of its CEO Sam Altman, with authorities discovering an anti-AI manifesto, a lighter, and kerosene at the scene. The incident forms part of a disturbing pattern of attacks and threats targeting the tech sector, as researchers warn that the industry’s rapid push into artificial intelligence is radicalizing segments of the public already frustrated by its environmental toll, safety lapses, and disregard for communities.

In April, Italian authorities arrested an Instagram influencer inspired by Ted Kaczynski for plotting anti-tech attacks. Two self-described ecofascists who carried out a deadly shooting at a San Diego mosque last month cited “AI slop” and political ties to tech firms like Palantir in their manifesto. An Indianapolis city councilor found a note reading “NO DATA CENTERS” after gunshots were fired into his home. These cases, documented across law enforcement reports and manifestos, signal how grievances over AI infrastructure and deployment are intersecting with existing extremist ideologies.

Jordyn Abrams, a researcher at George Washington University’s Program on Extremism, described AI as a novel driver of political violence. While earlier concerns centered on terrorists exploiting tools like ChatGPT, the focus has shifted to backlash against the companies themselves. Local opposition to data centers has grown beyond protests into direct action in some areas, fueled by the massive energy demands of these facilities.

A Business Insider analysis of permits shows US data center power consumption could rise 50 percent if all approved projects through 2025 come online, reaching between 224 and 358 terawatt-hours annually. That scale rivals the electricity use of most states outside Texas. Hyperscale centers, many built on rural farmland and woodlands, account for the bulk of new demand, with companies like Amazon planning vast complexes that reshape landscapes and strain local grids.

Critics argue these developments proceed with little meaningful oversight. The same breakneck rollout that alarms environmentalists and residents has also exposed dangerous shortcomings in AI applications. A survivor of a January 2025 school shooting in Nashville is suing Omnilert, the firm behind an AI gun detection system installed in the district’s schools. The lawsuit claims the technology failed due to camera angles, lighting, and other limitations the company had downplayed in its marketing. Two people died in the attack, and the suit alleges Omnilert knew or should have known its system could not reliably perform in real emergencies. School officials later acknowledged the footage was not close enough for the AI to trigger an alarm.

Meanwhile, major AI firms appear more focused on commercial lock-in than broad safety. OpenAI and Anthropic are shifting emphasis toward coding tools that generate vast usage and make customers dependent on proprietary codebases, according to industry observers. This approach prioritizes recurring revenue over addressing the societal disruptions their core technologies create.

Polling indicates widespread public anxiety about AI’s effects on jobs and daily life, yet skepticism remains high about government’s ability to keep pace. Some voices on the right have used that doubt to argue against any regulatory role, claiming state involvement would stifle innovation. Yet the absence of guardrails has allowed the industry’s expansion to outrun both environmental planning and basic accountability measures, leaving communities and even schools to absorb the consequences.

Extremist responses remain rare, but researchers note they emerge from the same reservoir of resentment building in town halls and online forums over data centers, energy costs, and unproven safety claims. The pattern suggests that without slower, more deliberate development and genuine public input, the tech sector’s ambitions will continue generating not just efficiency gains but also alienation and, at the margins, violence.