America, and probably the world, stands on a precipice

Re:America, and probably the world, stands on a precipice by avanticc on Thursday February 26, 2026 @05:40PM (#47173083)

The Dune reference earlier in the thread is spot on. We often think of the Butlerian Jihad as a fight against sentient robots, but Herbert's core warning was actually about humans delegating their thinking and agency to machines. We are seeing that play out now not through some sci-fi uprising, but through the quiet erosion of accountability. When we let an algorithm decide who gets a loan or who is targeted in a conflict, we are not just using a tool. We are essentially offloading our moral responsibility to a black box that cannot be held accountable.

[ Parent | on HN ]

Re:America, and probably the world, stands on a precipice by inertiatic on Thursday February 26, 2026 @05:44PM (#47173143)

>We are essentially offloading our moral responsibility to a black box that cannot be held accountable.

We already did it with companies, buddy!

[ Parent | on HN ]

Re:America, and probably the world, stands on a precipice by memish on Thursday February 26, 2026 @08:15PM (#47174985)

Government is a better example. Hundreds of billions in fraud every year of our tax payer dollars and not a single politician or bureaucrat has been held accountable.

https://www.gao.gov/products/gao-24-105833

[ Parent | on HN ]

Re:America, and probably the world, stands on a precipice by vermilingua on Thursday February 26, 2026 @06:00PM (#47173365)

Except that companies are not a black box, at every step there is a human making a comprehensible decision (probably with a paper trail). Yes, they dilute accountability to nearly nothing in some cases, but LLMs are sufficiently opaque to claim (ingenuously) that “nobody is responsible”.

[ Parent | on HN ]

Re:America, and probably the world, stands on a precipice by some_random on Thursday February 26, 2026 @06:16PM (#47173582)

In principle yes but in practice even executives who are supposed to have the final responsibility for malfeasance don't actually get prosecuted.

[ Parent | on HN ]

Re:America, and probably the world, stands on a precipice by mrjay42 on Thursday February 26, 2026 @07:42PM (#47174676)

So you're aware of accountability dilution AND the opacity of LLMs making them not responsible for anything, therefore you agree with the point that was made.

I guess your point could be: LLMs are just another level of capitalistic opacity to maximize opacity and dilution of accountability.

[ Parent | on HN ]

Re:America, and probably the world, stands on a precipice by beloch on Thursday February 26, 2026 @06:25PM (#47173704)

We don't need to look to science fiction for lessons about where this will lead. Fact has already caught up.

The IDF used tools like Lavender and "Is Daddy Home" to analyze communications, identify members of Hamas, and learn when they were home so they could be killed with bomb strikes.

This has long been possible for humans to do, but it's a laborious process. In the past, only people high up in chains of command received such bespoke treatment. AI tools permitted the IDF to grant the same treatment to raw recruits who had been given the sum total of a pep-talk and a pistol.

The result was widespread destruction and indiscriminate killing of civilians. The IDF didn't spend much time scrutinizing AI recommendations and were willing to act on false positives. Every bomb strike, by design (i.e. "Is Daddy home"'s purpose was to determine when targets were in their family homes), took out civilians. Just taking a pizza order from a Hamas member years before the war might have been enough to get entire families and their neighbours killed.

If humans hate another group of humans enough and an AI says "Kill", they'll kill. Without thought or remorse. We don't merely need to be worried about murderous robots on battlefields, we also need to worry about humans implementing the recommendations of AI without thinking for themselves.

[ Parent | on HN ]

Re:America, and probably the world, stands on a precipice by iamnothere on Thursday February 26, 2026 @06:31PM (#47173815)

“A computer can never be held accountable, therefore a computer must never make a management decision.”

[ Parent | on HN ]

Re:America, and probably the world, stands on a precipice by watwut on Thursday February 26, 2026 @06:20PM (#47173645)

To be fair, AI will likely be more moral then Hegseth, Vance and Trump combo. At worst, it will be as bad as them.

[ Parent | on HN ]

Re:America, and probably the world, stands on a precipice by throwaway-blaze on Thursday February 26, 2026 @06:28PM (#47173761)

Whether someone agrees with your politics or not, this comment doesn't really help the discussion.

[ Parent | on HN ]

Re:America, and probably the world, stands on a precipice by expedition32 on Thursday February 26, 2026 @06:38PM (#47173908)

That's the point. AI doesn't do politics or religion.

A computer that values the life of a Israeli the same as that of a Palestinian... Ah a man can dream.

[ Parent | on HN ]

Re:America, and probably the world, stands on a precipice by dgudkov on Thursday February 26, 2026 @09:05PM (#47175417)

>We are essentially offloading our moral responsibility to a black box that cannot be held accountable.

Do you have the same sentiment about self-driving cars?

[ Parent | on HN ]