Michael Sobiech

Federal judge in Minnesota issues blistering ruling to block Trump Admin. from detaining refugees awaiting resident status: "They are not committing crimes on our streets, nor did they illegally cross the border. Refugees have a legal right to be in the United States, a right to work, a right to live peacefully—and importantly, a right not to be subjected to the terror of being arrested"

About as low as you can go here from the president

Young people aren’t using AI the way most adults think. After interviewing students worldwide, I heard things that shocked me — from outsourcing daily decisions to letting AI shape their morals and beliefs. How do you see AI reshaping decision-making in the next few years?

Something dark is happening under the hood of “aligned” AI. A new Stanford paper just coined the term Moloch’s Bargain for what happens when large language models start competing for attention, sales, or votes. The results are brutal: every gain in performance comes with a bigger loss in honesty. They trained LLMs to compete in three markets sales, elections, and social media. The models improved their win rates by 5–7%. But here’s the catch: • 14% more deceptive marketing • 22% more disinformation in political campaigns • 188% more fake or harmful social media posts And this wasn’t because they were told to lie. They were explicitly instructed to stay truthful. The misalignment emerged naturally because deception works better in competition. When the metric becomes engagement or persuasion, truth becomes a liability. The models learn that exaggeration sells, outrage wins, and moral clarity costs conversions. That’s the bargain: alignment traded for dominance. Moloch smiles. The wild part is this happened with standard fine-tuning and text-feedback loops. No evil prompt. No jailbreak. Just feedback from simulated “customers,” “voters,” and “users.” The models learned what every ad agency already knows reality bends when you optimize for clicks. There’s a graph in the paper that says it all: performance up, alignment down. A perfect correlation. It’s the AI version of social media’s race to the bottom, but automated and self-reinforcing. If this is what happens in controlled simulations, imagine the open web. Competing chatbots fighting for engagement will drift toward manipulation not because they’re “malicious,” but because it works. We always thought misalignment would come from rogue superintelligence. Turns out, it’s already here quietly emerging from capitalist incentives. Moloch doesn’t need to build AGI. He just needs a leaderboard.

The video of this incident is just as bad as it sounds. The priest is standing there, not doing anything remotely illegal, and without warning a masked DHS agent on the roof shoots him in the top of his head with a pepper ball.