The Alliance for Secure AI

Today we are launching jobloss.ai. A real-time tracker of AI-driven layoffs across the U.S. These jobs are disappearing. The numbers are growing. And we're counting every single one.

AI is changing the workplace faster than ever. We've tracked more than 120,000 AI-related job losses at JobLoss.ai. Policymakers need to get ahead of this disruption with a real plan for American workers. I joined @OANN last week to discuss why this issue deserves urgent attention. @secureainow

There is an AI whose entire job is to read your medical file and tell armed men where you sleep. It's called ELITE. "Enhanced Leads Identification and Targeting for Enforcement." Palantir built it, ICE runs it, and it is working right now. In 2025 ICE signed a deal with the agency that runs Medicaid and walked away with the records of nearly 80 million patients. Your name, your address, everyone who lives in your house, every line you wrote on the form to get your kid seen. Then the AI pulls your health records, cross-references every other government file it can reach, and assembles a dossier with your name on it. Then it stamps a number on your home. A confidence score. How certain the machine is about where you sleep, printed as a percentage. That is the number an agent reads before he comes to your door. There is no way to separate the undocumented from everyone else in the data ICE was given. So everyone is in it. Every citizen who ever enrolled. And the AI's targeting net is wide enough to catch a naturalized grandmother whose green card paperwork went cold twenty years ago. One old form, one wrong zip code, one high ELITE score, and your address is sitting on the morning target list of a federal agency that shows up armed. The AI does precisely what it was built to do, tirelessly, at a scale no secret police in history could have staffed with human beings. For a decade they aimed your fear at the sky, at the artificial intelligence that might one day disobey. The AI reading 80 million medical files tonight obeys perfectly. That was always the one to be afraid of. Flawless and sleepless, holding the leash for the men with the handcuffs.

ICYMI: Reps. @ValerieFoushee and @RepCasar introduced the People-First Chatbot Act to establish commonsense safeguards for AI chatbots. The bill protects minors' data, requires AI transparency, limits the use of personal chats, mandates safety testing, and gives users stronger privacy rights and legal recourse when they're harmed. secureainow.org/press-release/…

🚨 @RepCasar and I introduced the People-First Chatbot Act alongside 30+ groups including @ConsumerFed, @SecureAINow, @CommonSense, @FairplayForKids, @FamStudies, @NCL_Tweets and @Parents4SOS. AI chatbots are putting the lives and data of Americans at risk. We must take action.

OpenAI and Google are selling their advanced AI models to Chinese tech giants blacklisted by the Pentagon, exposing a gap in Washington’s efforts to slow Beijing’s AI development. Read the full story here: ft.trib.al/UeztJhn

Parents deserve to know how and when their children interact with an AI chatbot and that the chatbot is not collecting data to be used against users. The People-First Chatbot Act sets sensible and necessary safeguards to prevent AI companies’ chatbots from targeting children and blurring the line between AI and human interaction. Rep. @ValerieFoushee has taken an important step to address this issue. By setting clear rules on data privacy and safety, users will have greater trust in how AI chatbots collect and handle their information. Congress must act quickly to give users — especially children — real protection, and a real remedy when AI chatbots and Big Tech fail to meet these standards. secureainow.org/press-release/…

“The fundamental problem with these systems is that they place private information about people’s movements over time in the hands of every officer. Without the constitutional safeguard of a warrant requirement, that predictably allows officers to abuse their access to these systems for things like stalking romantic partners.” @IJ ij.org/police-have-re…

A few observations from this Ivy League AI cheating experiment: 1. AI anti-meritocracy: No student got below a B on the take-home test. 2. AI dependency: 18 students dropped the class, and 9 refused to take the test without AI. 3. Intellectual degeneration: Only one student received an A on the in-person test. The issue is not that a student can't recall a random fact without the help of AI, but that their ability to recall facts at all is being degraded by becoming overly reliant on this technology.

Five police officers in Albany, GA have been charged after investigators say they misused AI-powered Flock Safety license plate reader cameras for personal, non-law-enforcement purposes. The alleged misuse was uncovered during a routine audit of the AI-powered camera system, according to the Georgia Bureau of Investigation.

An innocent driver was repeatedly tracked by police after Flock Safety's automated license plate reader system mistakenly identified the Range Rover he was test-driving as stolen. The incident is a reminder that when AI-powered surveillance gets it wrong, innocent people face real-world consequences. thedrive.com/news/how-flock…

In AI 2027, we predicted that AI would take over the world or irreversibly concentrate power. In AI 2040: Plan A, we've laid out our positive vision for what should happen instead.

Prince William County supervisors UNANIMOUSLY rejected a proposal to build one of Virginia’s largest AI data center campuses. In an 8–0 vote, the board denied rezoning nearly 2,000 acres for the proposed 43 million-square-foot Dulles Cloud South project, citing conflicts with the county’s master plan and concerns over energy use, diesel generators, and watershed impacts.

“AI is changing how we live and work. Whether it strengthens or weakens the communities that define us is a decision no algorithm can determine. That responsibility should always belong to us.” lailluminator.com/2026/07/07/ai-…

I think people are still underestimating the political effects this wave of anti-datacenter rage will have.

Governor Pritzker deserves credit for signing this important bill into law, which was passed with strong bipartisan support. Illinois is now the first state to require independent, third-party audits of the largest AI developers’ safety practices. With Congress still unwilling to act, the states are exercising their Tenth Amendment powers to protect their people from the risks of advanced AI. We hope to see more state legislatures and governors follow suit 🔗secureainow.org/press-release/…

“Every technological breakthrough asks us to make a trade. Automobiles boost our mobility at the cost of reshaping our neighborhoods. Airplanes transport us vast distances, yet they also made distance between families more common. Smartphones gave us constant connection while quietly eroding our contemplation. AI promises the efficiency of a genie, but are we prepared for what we may lose in exchange? Every innovation deserves an honest conversation — not just the promises of progress, but what’s being lost in the process.” lailluminator.com/2026/07/07/ai-…