AI can't be trusted

[WillyAp]

The AI can't be trusted.
Morgan Stanley seeks security in non AI affected Assets, like Bitcoin, Crypto, Gold, Silver, mines etc.

Morgan Stanley Investment Management is adding to emerging-market bets that are insulated from the artificial intelligence boom as it positions for a potential pullback in the crowded tech trade, according to Deputy Chief Investment Officer Jitania Kandhari.
Rising concern over stretched valuations and heavy capital spending by US technology giants has pushed the Nasdaq 100 Index to its first loss-making start to a year since 2022. The unease is echoing across developing markets, where nearly half of a $12 trillion rally since April has been concentrated in AI-linked stocks.

https://www.bloomberg.com/news/articles/2026-02-23/morgan-stanley-im-adds-emerging-market-assets-without-ai-risks
no paywal: https://archive.ph/5fdzP

The US War Ministry pulled away from Anthropic saying the the AI wished to spy on every US citizen.
While CEO Dario Amodei’s decision to refuse the Pentagon’s demand to allow Claude to kill people autonomously is a bit hilarious as war implicates killing very much.

https://www.siliconvalley.com/2026/02/27/anthropic-sees-support-from-other-tech-workers-in-feud-with-pentagon/

while

AI's can’t stop recommending nuclear strikes in war game simulations

Leading AIs from OpenAI, Anthropic and Google opted to use nuclear weapons in simulated war games in 95 per cent of cases

https://www.newscientist.com/article/2516885-ais-cant-stop-recommending-nuclear-strikes-in-war-game-simulations/

The more is known of AI or the motivation of the leadership the clearer ist becomes to stay away.

[Ucy]

Has no feeling or fear, so will most likely be psychopathic in behavior unless you feed it correctly with the word of the CREATOR of the Universe, then it would be more caring for life, fear the CREATOR and never do as it pleases it. For now, it seems to do the will of their creators who likely believe in evolution theory which is based on the assumption that Earth and life aren't that unique or important... Things that believe in the theory should never be trusted.

As for Bitcoin, it's rule by its own unique non-AI program that is far more advanced than AI could ever be, even if you combined all AI processors in existence or in the future

[WillyAp]

Has no feeling or fear, so will most likely be psychopathic in behavior unless you feed it correctly with the word of the CREATOR of the Universe, then it would be more caring for life,

AI is developed by people with little love for anything but themselves.
Most people pressed into jobs care just for the Weekend.

[0xredhot]

I think it cannot be stopped biggest crypto wallets out there are using AI to build their products. So from now on software wallets that many people use are build with AI.

[Veenie]

I think it cannot be stopped biggest crypto wallets out there are using AI to build their products. So from now on software wallets that many people use are build with AI.

Biggest crypto wallets use Ai to build their products? You might want to do a little research on that because some companies that claim their product is "written entirely by Ai" have been ousted as facetious in a bid to sell that Ai pipe dream.

And there's been a ruling in the US recently that generative Ai cannot be copyrighted, granted it's not trickled down to other countries but that act alone? How could a product entirely built with Ai scale and bring in profits? If every Tom, Dick and Harry can scrape off their code and be a better marketer and since the code can't be copyrighted, there's no inherent legal consequence.

[0xredhot]

I think it cannot be stopped biggest crypto wallets out there are using AI to build their products. So from now on software wallets that many people use are build with AI.

Biggest crypto wallets use Ai to build their products? You might want to do a little research on that because some companies that claim their product is "written entirely by Ai" have been ousted as facetious in a bid to sell that Ai pipe dream.

And there's been a ruling in the US recently that generative Ai cannot be copyrighted, granted it's not trickled down to other countries but that act alone? How could a product entirely built with Ai scale and bring in profits? If every Tom, Dick and Harry can scrape off their code and be a better marketer and since the code can't be copyrighted, there's no inherent legal consequence.

I already did, that is why I said what I said, If you think you cannot build product entirely with AI and bring in profits you havent used claude I guess, problem is not building problem is distribution in the end always.

[Tonia86]

AI is often most useful where we're already expert enough to spot its mistakes, yet least helpful in the deep work that made us experts in the first place. It works best for tasks we could do ourselves. .

AI cannot be fully trusted for critical, unverified tasks due to risks of misinformation, "hallucinations," and bias.

[WillyAp]

AI is often most useful where we're already expert enough to spot its mistakes, yet least helpful in the deep work that made us experts in the first place. It works best for tasks we could do ourselves. .

AI cannot be fully trusted for critical, unverified tasks due to risks of misinformation, "hallucinations," and bias.

You need to check even the date. It causes more work, not less.

[Vod]

The AI can't be trusted.

The arguments you are making is that humans cannot be trusted.    AI is just a tool that we are improving - it will be humanity that ultimately presses a key that destroys everything. 

[PrivacyG]

The AI can't be trusted.

The arguments you are making is that humans cannot be trusted.    AI is just a tool that we are improving - it will be humanity that ultimately presses a key that destroys everything. 

It depends.  If the Artificial Intelligence is coded to 'think' for it self, with things it calculates and randomness it can be a thing the Artificial Intelligence triggers on its own.  Here is the problem I see, they are trying to make it 'think'.  And in many ways, it would be stupid and useless to make it pretend it has feelings.  In fact.  It may be even dangerous.  But thinking with out any feeling is very dangerous too.

[WillyAp]

The AI can't be trusted.

The arguments you are making is that humans cannot be trusted.    AI is just a tool that we are improving - it will be humanity that ultimately presses a key that destroys everything. 

With an error rate of above 50% no one would even employ someone like that.
Hardly anyone of the hypers takes that into account.