GW Researcher Ties 2026 Anti-Tech Attacks to Trust Collapse

An anti-AI movement that until recently looked like online venting is now showing up at front doors. In a Tech Policy Press essay published June 22, Jordyn Abrams, a research fellow at George Washington University's Program on Extremism, walks through three 2026 incidents that together suggest a pattern rather than isolated cranks. On April 6, a shooter fired 13 rounds at the Indianapolis home of Councilman Ron Gibson, who had backed a contested data center project, and left a note reading 'NO DATA CENTERS.' Days later, Daniel Moreno-Gama allegedly threw a Molotov cocktail at OpenAI CEO Sam Altman's residence and tried to breach the company's headquarters, reportedly carrying a manifesto naming AI leaders as targets. In June, the FBI disrupted a plot involving drones with explosives aimed at a UFC event at the White House, with suspects who, per Abrams, were concerned about 'data centers taking up all the water in communities.'

Abrams's argument is that this is a trust problem before it is a security problem. She cites polling that only 17 percent of Americans say they trusted the US government to do what is right, that 31 percent trust the government to regulate AI responsibly, and that just 15 percent trust AI companies themselves with development decisions. Sitting on top of that is a Washington that looks, to a skeptical voter, fused with the industry it is supposed to police. Tech companies, she notes, spent a combined $50 million on federal lobbying over nine months heading into the 2026 midterms. Elon Musk runs the Department of Government Efficiency. Trump's May 2026 China trip included Musk, Tim Cook and Jensen Huang. The Department of Justice has argued in court that xAI is 'vital' to national security.

The policy prescription is more political than technical: democratic engagement with the concerns driving opposition, intelligence work that targets violent intent rather than broad ideology, and visible separation between officials and tech elites. Abrams points to 14 states weighing data center moratoriums and to a June 2026 Trump executive order plus bipartisan House legislation as early signs that the system can route grievance through normal politics if it chooses to.

The honest caveat is that this is a single op-ed by a counter-extremism researcher, not an FBI bulletin, and Abrams does not claim the three incidents are operationally linked or sketch what 'targeted intelligence' would look like in practice without scooping up nonviolent critics. The reporting also does not pin down what the June executive order or the House bill would actually require of AI firms or data center operators, which is where the next real test sits. For practitioners, the forward-looking read is simpler: 'policymakers must meet their constituents where they are on AI,' and the firms that get out ahead of that conversation, on water, power, jobs and siting, are the ones least likely to find their executives' addresses in the next manifesto.