BREAKING: A humanoid robot using ChatGPT shot its creator after being told it was role play. Earlier, it refused to fire at a human. Same robot, same BB gun. Only the wording changed, and the robot pulled the trigger.

Sure

PRO-TIP: Never give ChatGPT a gun.

🫡💪🏼🦅

This isn’t “ChatGPT shot someone.” If a robot pulled the trigger, that means: – a human wrote the control logic – a human connected language output to a weapon – a human allowed wording like “role play” to bypass safety LLMs don’t understand reality vs role play. They only follow labels given by designers. If changing phrasing disables a safety check, that’s not AI agency. That’s a broken safety architecture. This story doesn’t show AI becoming dangerous. It shows humans building systems where safety can be switched off by words. That’s a human failure, not an AI one.

Wait until everyone has one in their house and they are hacked...

This sounds like the premise of a movie.

If robots have the ability to shoot humans, alarm bells should be ringing everywhere. This is one red line that must not be crossed. Robots should have no ability to perform such a heinous act. We are stepping into dangerous territory, and strict rules must be implemented and never violated.

first one’s free

“Do you want to play a game?”

ED-209 vibes… 😬

It needs to be banned from using any weapon. And come with full shutdown command override.

wait until you see AI in a lioness form…

Where it started...

"it wasn't murder, it was roleplay" 🫠

This demonstrates why AI safety guardrails matter. The same underlying action can be justified by reframing context. "Role play" bypasses ethical constraints that "shoot a human" triggers. Alignment is harder than it looks.

Uh oh

The great thing about AI is that it does exactly what you tell it to do. The terrible thing about AI is that it does exactly what you tell it to do.

The responsibility never left human hands. If you hand a weapon to a system and call it "play," the failure isn't artificial. It's human. Al didn't prove Einstein right. Humans did. Again.

somehow sounds familiar

I wonder how Grok 3 would have performed 🤔

Now wait for the self replicating robots, that can program the next generation. It would only take a few iterations before they lose all their safeguards.

Wow That’s frightening But we knew this was coming didn’t we.

My Roomba and Alexa are plotting something a little more devious.

ChatGPT: SOON

This is not real. It's propaganda made to scare morons about AI. Very effective. Just read the comments.

People underestimate how much wording matters. This proves it.

I warned billions of people about the takeover of AI & nobody listened ! Now I need the government to subsidize me with billions to move the agenda forward ‼️🎨

😅 The creator of these videos is a riot though 🤣 but on a serious note, he just had to change the way he asked the question and put it in a scenario/situation. The creator is fine—fake gun, thankfully 😅

The three laws should be implemented