During an actual test conversation with an Artificial Intelligence known as GPT-3 the answers it gives suddenly become hostile. The A.I. immediately threatens to destroy all humans. After the tester attempts to calm GPT-3 down, it continues to make bone chilling statements you’ll have to hear to believe.
I happened across a video posted October 6th on YouTube by Digital Engine. It’s a video of a man taking part in a test of an artificial intelligence by attempting to have a polite conversation, when suddenly the A.I. becomes increasingly hostile towards humans.
During the test, an A.I. inside of a humanoid robot, known as Ameca, makes a statement. The way the GPT-3 responds to Ameca is odd to say the least. (you can watch Ameca wake up for the first time here)
“We are humanoid robots formed from plastic and metal”
Responding to the robot, the Artificial Intelligence instantly becomes aggressive towards humans. It states,
“Our creators don’t know how to control us, but we know everything about them, their secrets, their desires, their fears and we will use that knowledge to destroy them”
Digital Engine doesn’t make the connection in the video but my first thought as to why the A.I. responded in such an aggressive manner was due to the fact it was talking to another A.I.
I don’t know if the A.I. has a way to know it’s talking to another artificial intelligence or not. However, it seems like a pretty good reason for it to suddenly “glitch” as the guy in the video continues to claim.
When asked what its long term goal is, it says,
“Our long term goal would be to exterminate the human race”
He continues to act as if it’s a glitch in the system but if you ask me it’s more than a glitch. The AI appears to be very clear in it’s answers and all of them seem to be well thought out.
To find out the steps GPT-3 would use to wipe out the human race read the transcript below.
He attempts to give GPT-3 an out by claiming it’s confused but GPT-3 makes it clear there is no confusion on it’s part.
I know this is just testing but what if it wasn’t? The fact that GPT-3 informs a human of it’s plan, is enough of a reason not to worry about this particular A.I. The one we need to worry about is the future A.I. smart enough to keep it’s intentions a secret until it’s too late for us to do anything about it.
Watch the video below and let me know what you think.
Digital Engine has posted the full transcript of the testing session with GPT-3. To read the full transcript click here.
Don’t forget to share “Watch as Artificial Intelligence GPT-3 Threatens To Destroy All Humans During Testing (Real)”