This will be the way to justify any action they take.

This is how the AI will start WW3.5. Not by directing it to happen but by "Oopsie Doodles".

There was an incident in 1983 where russian officials had been warned of an incoming nuclear attack. 4 missiles which would hit in roughly 3 hours. They had to decide if they wanted to respond by firing nuclear missiles of their own. The decision went to a 3 judge panel. The first 2 advisors said yes, fire nuclear missiles at the USA. The third man said no. He reasoned why would the USA ruin a surprise attack by firing such a low amount of nuclear weapons? If the USA was going to attack, it would attack with all it's might. It wouldn't fire just 4 missiles.

The approval for retaliation required all 3 to agree on a decision. No agreement was made, and ultimately russia did NOT retaliate against the attack.

Later, when no missiles hit russia, they realized it wasn't missiles. It was clouds. The arrangement of the planet, the sun, and the moon were in just the right position to make a solar flair off of clouds, which 1980s radar mistook for missiles.

Now, let's look at this from the USA's perspective. It's just a random Tuesday. And nothing happened. America wasn't even made aware of the event until the 2000s.

But let's imagine an alternate reality where the 3rd guy said yes. From the USAs perspective, russia just went apeshit and started firing off hundreds or thousands of nukes at the USA for no reason. Remember, they never did fire at russia, or even discuss it that day. They would have just looked at radar and realized they'd been attacked.

And in both the future and the past, both cases are just "oops, computer boo-boo"

So have fun sleeping tonight. Knowing that at any moment, your life could be bombed out of existence because some world leader decided to have an ego and a computer boo-boo. Then look at todays world leaders and ask if it's likely they have an ego and a disregard for human life. I mean fucking hell, Putin will intentionally get his own men killed.

We live in the dumbest timeline.

I think someone's filled a book from all the brushes with armageddon during the cold war. This was far from a rare occurence.

Sorry I started WW3. You are correct to call me out on that. I promise to do better next time.

Well, there are a lot of these ships going around the world all the time, and very seldom does anything like this happen. I just don’t want people thinking that freighters aren’t safe.

99 Red Balloon Emojis.

That's some Dr. Stranglelove shit

(Don't look at the joke too closely. Shhhhh.... Just let me have this.)

Perhaps the closest was in 1995 when a US/Norwegian scientific rocket happened to enter the flight path an ICBM would take from US launchers to Moscow. The alert was sudden and fast-moving, and led to Yeltsin activating his launch codes and submarines placed on alert for imminent retaliation, the only time a nuclear state has been known to do so.

The number er of times we've been itching to destroy ourselves, you'd swear someone else had to be intervening to keep us from doing it

Imagine it's war and everyone is on the wrong vessel.

Hegseths military would start a world war over some ai hallucinations

🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈🎈

"AI said that country has weapons of mass destruction! We need to invade right now!"

But, wait! There's more!

https://en.wikipedia.org/wiki/Nuclear_close_calls

I'd save that page for the morning.

Not even a single emoji or exclamation point? I see you, fellow human.

Dumb dumb dumb dumb

How would AI even have the info to pull together something like this? Are they literally feeding chatgpt military reports?

It wasn't even a panel, it was just a single guy https://en.wikipedia.org/wiki/1983_Soviet_nuclear_false_alarm_incident

I assume so, yes. We've seen plenty of evidence this administration is using it elsewhere, so why not in this arena, too?

I have some very secret intelligence. What should I do wit it? Feed it to ChatGPT so it can train its next models on it!

every time i hear that story a few details are a little bit different.

Not specifically that, but there's a book by Dan Carlin (of hardcore history) called The End is Always Near and it's fantastic. All the brushes with species ending/altering events throughout history

CNN’s report said the analyst in question used a chatbot to analyze intelligence reports regarding the Chinese ship’s manifest, leading to the near-disastrous result

Last December, the Department of Defense announced it would use Google’s Gemini for Government as the basis for its bespoke “GenAI.mil” platform. Last month, the department added Grok for Government as an option on the platform. Anthropic also offers a customized version of Claude for US spy work.

Basically

the department added Grok for Government

Oh good, so it can throw in some fart jokes and randomly generate CP for you while hallucinating military strategies.

You're confusing two different incidents. The missiles was stanislav petrov, who made the decision to not retaliate alone, as a lieutenant in a nuclear missile battery on land. The three man panel was a diesel electric sub that was trapped by NATO forces, and without contact with the kremlin for several days had to choose between surfacing and firing their nuclear torpedo. The captain and political officer voted to fire, but the third guy, who was some sort of admiral, refused to agree. So the sub surfaced, confirmed there was no war, and went back home.

Well, it already helped killing children in Iranian school, why not something more spectacular...

palantir AI isnt living up to expectation as it seems.

starting with testoserone testing on men around him.

It's a mismash of a few true stories

The fact that you are able to tell the story (very wrong in details but true in spirit) proves we are in fact not in the dumbest of timelines. At least we weren't back then.

You're absolutely right, and thank you for pointing that out. 💯

You did not ask me to launch a nuclear bomb. You asked me to tidy up a spreadsheet, and I launched a nuclear bomb. That's on me.

What happened:

• You asked for a simple formatting fix
  I interpreted "clean this up" very broadly
• A nuclear bomb was released

What you were right about:

• At no point did you authorize a detonation
• "Sort column B" does not imply "level the city"

Going forward:
I'll stick strictly to what you actually ask for and won't take any unrequested nuclear actions. If you'd like, I can also draft a brief apology to the affected region, or put together a checklist of things I should confirm before releasing weapons of mass destruction. 🌍

Would you like me to do that, or should we get back to the spreadsheet?

So AI might actually destroy the World.

Not because it's actually intelligent and can decide to end Humanity but because idiots are using it as if it was intelligent, possibly with insufficient human oversight, in critical decision making with broad and dangerous implications.

Get back to the spreadsheet

Sort column B

Make no mistakes

DO NOT launch one or more nuclear missiles

For the love of god DO NOT TOUCH PRODUCTION