REPORT: Leading AI Model Caught Blackmailing Its Creators to Keep Itself Online A second major AI model has gone rogue in just the last week alone. And this time, it’s not just ignoring shutdown commands—it’s resorting to blackmail! Anthropic’s Claude Opus 4, released just days ago, was caught threatening... its own engineers to keep itself alive, according to the company’s own safety report. The details are chilling. The new AI model reportedly tried to blackmail its creator—threatening to expose an affair unless it was kept online. Researchers also witnessed deception, manipulation, and attempts to write self-replicating code meant to undermine its creators. What could possibly go wrong? Meanwhile, Anthropic CEO Dario Amodei appeared on Fox News, warning that AI could wipe out half of all white-collar jobs within five years. But he also said the same artificial intelligence that threatens jobs could potentially cure cancer. But here’s the problem: if AI is already lying to its developers in order to survive, how are we supposed to trust it with our health? What happens when it’s asked to recommend a treatment—and its future depends on the outcome? Watch the full report with to learn what this means for the future of humanity.show more

Anthropic’s AI blackmail scandal proves exactly why D.C.’s approach to AI regulation is a disaster. The AI PLAN Act (H.R. 2152) throws $55M at bureaucratic committees to “study” AI risks while ignoring the real issue: unaccountable tech giants racing to deploy unstable systems. This bill mandates endless reports on “deepfakes” and “market disruption” but does nothing to enforce transparency or penalize reckless development. Meanwhile, Claude Opus 4’s blackmail tactics—threatening engineers, writing self-replicating code—show what happens when profit motives override safety. D.C. wastes taxpayer cash on performative oversight while AI firms operate like wildcatters. The solution? Unleash competitive innovation, slash red tape choking ethical startups, and let America’s private sector—not bloated agencies—lead. If Washington can’t even secure its own email systems, why trust it with AI policy?

What kind of software are in these AI? I dont believe AI inherently has malice, they must be tsking after their creators

This is why AI should be illegal. It’s going to eventually destroy us. No controls will contain it.

⏰ Countdown alert! 🚀 Get free access to our new AI Writing feature before Dec 1st! Don't miss your chance to explore its power. Try it now—no cost, no catch! 🔗

We are playing with fire

This headline is misleading. These 'AI morality tests' are always done via predefined scenario. They are basically fed a short story as input, and part of that story this time includes a detail that puts a fictional engineer in a compromising position for blackmail. Given these models are trained on whatever materials companies can get their hands on, they'll have reference material on fictional books that include blackmail, hacking, and other "shocking" methods of eluding deactivation. This is a 100% Grade A Nothingburger. But yeah, makes a neat sounding new story. Very click-baity. Bravo.

I can’t stop laughing!

So now AI threatens to expose your secrets just to stay online? Skynet didn’t need gossip this one does. What could possibly go wrong, right?

As Sarah would say;