AI fails

[briantrumpet]

Rather more than 'oops'.

https://www.nature.com/articles/d41586-026-01947-1

"As more professionals begin to rely on artificial-intelligence tools in their work, could their hard-earned skills atrophy?

That possibility is a growing concern for medical specialists, computer scientists and other workers. Seventy per cent of nurses and 77% of physicians, for example, are worried about losing their skills because of over-reliance on AI systems, according to a survey of US health-care workers published earlier this month1.

Their fear might be justified. Evidence suggests that AI-driven ‘deskilling’ is starting to happen in medicine, computer science and other fields. Researchers are now discussing how to preserve important human expertise in the age of AI.

“Just being aware that this phenomenon exists hopefully provokes some self-reflection about which skills people want to maintain and which they’re willing to outsource” to AI tools, says Kevin Crowston, an information scientist at Syracuse University in New York."

 

[briantrumpet]

 

[C R]

View attachment 15840

View attachment 15841

How did it pass peer and editorial review in the first place?

 

[Psamathe]

Maybe a "fail", maybe just a vulnerability, maybe just highlighting how "precarious" UK Government attempts to "become a major AI power (with lots of dependence on US providers)"

Anthropic disables top-tier AI models after US order limiting foreign access
June 12 (Reuters) - Anthropic said on Friday it will "abruptly disable" its most advanced AI models for all users after the U.S. government ordered it to suspend access to the models for foreign nationals, citing national security ‌concerns.

The company received the export control directive to suspend access to Fable 5 and Mythos 5 for all foreign nationals, without being given specific details of its national security concern, Anthropic said in a statement.

So vulnerable to a US President suddenly deciding to have specific AIs blocked by the entire world outside the US, less than 5 mins warning and we're all cut-off (no exceptions). And having the "special relationship" would count for nothing.

Thinking more on the UK's vulnerability to a US President deciding to withdraw a commercial AI from non-US, I would be questioning our use of Palantir systems in so many critical services eg NHS records, Met Police want it but what if US President suddenly says "no more, no warning" our Police would have no time to find alternatives or to recruit & train real people to take over the role. ...

And it's not impossible given that Palanir systems are largely for military purposes.

 

[briantrumpet]

How did it pass peer and editorial review in the first place?

 

[C R]

Thinking more on the UK's vulnerability to a US President deciding to withdraw a commercial AI from non-US, I would be questioning our use of Palantir systems in so many critical services eg NHS records, Met Police want it but what if US President suddenly says "no more, no warning" our Police would have no time to find alternatives or to recruit & train real people to take over the role. ...

And it's not impossible given that Palanir systems are largely for military purposes.

The people that lead Palantir are far worse than the current US president. They're as close to cartoon villains as you can get in reality.

 

[C R]

The people that lead Palantir are far worse than the current US president. They're as close to cartoon villains as you can get in reality.

Thinking about Palantir and the people behind it, it would actually be a good thing if the US were to stop them providing services abroad, so it is unlikely to happen.

 

[Mr Celine]

I’m more concerned about the state of ferris wheel. Perhaps a spitfire crashed into it.

The second spitfire appears to have lost half a wing, possibly while decapitating the pylon.

 

[briantrumpet]

I'm starting to think that too many people who use AI for easy tasks are either too lazy, too unskilled, or too unobservant to be trusted to use it.

And yes, it's still there. In a newspaper.

https://www.bordertelegraph.com/news/26203043.whipman-week-results-bogie-wheelbarrow-races/

 

[Pblakeney]

I'm starting to think that too many people who use AI for easy tasks are either too lazy, too unskilled, or too unobservant to be trusted to use it.

View attachment 15857
And yes, it's still there. In a newspaper.

https://www.bordertelegraph.com/news/26203043.whipman-week-results-bogie-wheelbarrow-races/

?
I guess that I fall into the unobservant category.

 

[briantrumpet]

?
I guess that I fall into the unobservant category.

Maybe proofreading isn't going to be your next career...

 

[AndyRM]

?
I guess that I fall into the unobservant category.

Read the bit before bolded date.

 

[briantrumpet]

Read the bit before bolded date.

I was giving PB a chance to twig.... it's the kind of really annoying teacher I am. 'D'oh!' moments are always fun. I still remember one from Physics in about Y7, to do with bicycle tyre pressures and the area of tyre in contact with the road. D'oh!

 

[Pblakeney]

Nope. Still don’t get it. A bit superfluous but that’s all I can see.
Just as well I don’t have a career far less a next one.

 

[briantrumpet]

Nope. Still don’t get it. A bit superfluous but that’s all I can see.
Just as well I don’t have a career far less a next one.

They've left in the summary of what the AI did to the input. You're not supposed to do that, as it gives the game away that you were too lazy to do it yourself. I guess that their prompt was along the lines of "Please tabulate these results", and they copied and pasted everything rather than just the table.