Comment Looking forward to being 18 again. (Score 1) 1
Hey, I have the same chance at an AI Hallucination as anyone else!
Comment Re: Where's the "none" option? (Score 1) 78
I ignore Amazon's AI summaries, it gets it wrong half the time. As for Google, I may glance at the summary but, again, the AI usually gives me a lowest common denominator response and not the specific response I needed. So I skip past the AI giving me the metaphorical equivalent of a "I didn't read the book" book report, and go to the real responses. Yes, some of them may be AI generated, which is why I read more than one result as well.
Comment I used ChatGPT once, does that count? (Score 1) 78
I avoid using them for anything, but there was this one time two years ago I needed to know the name of the English Law from the sixteenth century that set up the Elizabethan apprenticeship system. I needed to know because coincidentally this was the main way you gained the closest thing to what we would call citizenship, in England at the time. All the books I could find mentioned this law with description only, seemed to assume everyonealready knew the name. ChatGPT knew, correctly, that it was called the Statute of Artificers. Other than that, I do my own research and writing in the traditional scholarly way.
Submission + - Musk's Reports of Social Security Payments to the Dead Are Greatly Exaggerated 1
theodp writes: While Doge.gov still vows to get to the bottom of an Elon Musk tweet claiming that "there are a lot of vampires collecting Social Security," the AP and others are reporting that Musk's reports of social security payments being made to dead people are greatly exaggerated.
"The Trump administration is falsely claiming that tens of millions of dead people over 100 years old are receiving Social Security payments," reports the AP. "It is true that improper payments have been made, including some to dead people. But the numbers thrown out by Musk and the White House are overstated and misrepresent Social Security data. [...] A series of reports from the Social Security Administration’s inspector general in March 2023 and July 2024 state that the agency has not established a new system to properly annotate death information in its database, which included roughly 18.9 million Social Security numbers of people born in 1920 or earlier but were not marked as deceased. This does not mean, however, that these individuals were receiving benefits. The agency decided not to update the database because of the cost to do so, which would run upward of $9 million."
"Know Thy Data," AnnMaria De Mars wrote in a 2016 blog post, "[is] the most important commandment in statistics. [...] It is crucial to understand how your data are coded before you go making stupid statements like the average mother is 3 months old." While it was offered for the likes of her epidemiology students, De Mars' advice would also be well-heeded by the richest person in the world as plays data scientist with the nation's data.
"The Trump administration is falsely claiming that tens of millions of dead people over 100 years old are receiving Social Security payments," reports the AP. "It is true that improper payments have been made, including some to dead people. But the numbers thrown out by Musk and the White House are overstated and misrepresent Social Security data. [...] A series of reports from the Social Security Administration’s inspector general in March 2023 and July 2024 state that the agency has not established a new system to properly annotate death information in its database, which included roughly 18.9 million Social Security numbers of people born in 1920 or earlier but were not marked as deceased. This does not mean, however, that these individuals were receiving benefits. The agency decided not to update the database because of the cost to do so, which would run upward of $9 million."
"Know Thy Data," AnnMaria De Mars wrote in a 2016 blog post, "[is] the most important commandment in statistics. [...] It is crucial to understand how your data are coded before you go making stupid statements like the average mother is 3 months old." While it was offered for the likes of her epidemiology students, De Mars' advice would also be well-heeded by the richest person in the world as plays data scientist with the nation's data.
Submission + - NOAA scientists need permission to speak to external colleagues now (www.cbc.ca)
sandbagger writes: Travelling for international meetings or even joining a call with Canadian counterparts has become impossible for some U.S. government scientists, under new directives since U.S. President Donald Trump took office.
Submission + - Google will use AI to estimate a user's age (acs.org.au) 1
haxmor writes: Google says it will begin using artificial intelligence systems to estimate the age of its users, as it responds to potential changes to online safety legislation in countries such as Australia and the United States.
Submission + - Is Confirmation Bias Driving Elon Musk to Jump to Sensationalistic Conclusions? 2
theodp writes: "According to the Social Security database," said Dept. of Government Efficiency (DOGE) Chief Elon Musk in a tweet that's been viewed 54.2+ million times (and counting), "these are the numbers of people in each age bucket with the death field set to FALSE! Maybe Twilight is real and there are a lot of vampires collecting Social Security." In lieu of presenting a thoughtful, evidenced-based audit report, Musk made his case with a couple of ROTFL icons and a single screenshot of a simple table that seems designed to suggest nearly 400 million people — more than the population of the United States — are receiving Social Security benefits. That "there are FAR more 'eligible' social security numbers than there are citizens in the USA," Musk added, "might be the biggest fraud in history." Never mind that Social Security Beneficiary Statistics published by the SSA are much lower than those floated without explanation by Musk. Or that non-citizens including students and workers are also issued Social Security Cards (including Tesla's non-citizen workers) and can be entitled to benefits.
There are undoubtedly big fraud problems to solve at Social Security, just as there are at Musk-founded PayPal and other organizations. But Musk may want to take steps to help ensure that DOGE's work is driven more by informed consideration of evidence and less by jumping to conclusions based on confirmation bias. "Narrowly pursuing an investigation into what you initially suspect doesn’t just trip up scientists," warns 5 Ways Auditors Can Overcome Confirmation Bias. "Confirmation bias—one of five common judgment biases—has the potential to lead auditors up the wrong path just as easily. [...] The deeper one gets into investigating a particular hypothesis, the more difficult it becomes to consider contradictory ones. Rather, it’s common to seek evidence that supports suspicions and overlook data that don’t. Result: You’ve confirmed your bias—bypassing both the scientific method and best practices in auditing."
Writing about the resignation of acting SSA Commissioner Michelle King after members of Musk's DOGE team sought access to the agency's data, the New York Times reports that Martin O’Malley, who served as commissioner of the SSA in the Biden administration, said the claims of Musk and his team about the agency were not true. “They’re just making” things up, he said, referring to Musk’s suggestion that more than a million people in the Social Security database are in the 150 to 159 age range.
There are undoubtedly big fraud problems to solve at Social Security, just as there are at Musk-founded PayPal and other organizations. But Musk may want to take steps to help ensure that DOGE's work is driven more by informed consideration of evidence and less by jumping to conclusions based on confirmation bias. "Narrowly pursuing an investigation into what you initially suspect doesn’t just trip up scientists," warns 5 Ways Auditors Can Overcome Confirmation Bias. "Confirmation bias—one of five common judgment biases—has the potential to lead auditors up the wrong path just as easily. [...] The deeper one gets into investigating a particular hypothesis, the more difficult it becomes to consider contradictory ones. Rather, it’s common to seek evidence that supports suspicions and overlook data that don’t. Result: You’ve confirmed your bias—bypassing both the scientific method and best practices in auditing."
Writing about the resignation of acting SSA Commissioner Michelle King after members of Musk's DOGE team sought access to the agency's data, the New York Times reports that Martin O’Malley, who served as commissioner of the SSA in the Biden administration, said the claims of Musk and his team about the agency were not true. “They’re just making” things up, he said, referring to Musk’s suggestion that more than a million people in the Social Security database are in the 150 to 159 age range.
Submission + - TikTok Ban Linked to Pro-Palestine Content, Not China Threat (middleeasteye.net) 1
hackingbear writes: The main reason behind the United States' push to ban social media application TikTok is due to Israel’s image rather than fears of Chinese infiltrations, US Senator Mark Warner and Mike Gallagher have revealed during a panel the Munich Security Conference. Warner, the top Democrat on the intelligence committee, introduced the TikTok ban bill in 2023 along with Gallagher,who is the current Palantir executive. “So we had a bipartisan consensus,” Gallagher said. “We had the executive branch, but the bill was still dead until October 7th. And people started to see a bunch of antisemitic content on the platform and our bill had legs again.” A memo produced by the State Department for its Near East Affairs diplomats, which Klippenstein obtained, describes how Israel's deputy director general for public diplomacy at the foreign ministry, Emmanuel Nahshon, blamed the youth’s opposition to the war on Gaza on TikTok’s algorithm. The memo added that Nahshon said the youth’s public opinion was shifting because “the Tik-Tok algorithm favours pro-Palestinian content”. Gallagher also said in Munich that TikTok had made a “huge miscalculation” in its attempt to circumvent the ban. When TikTok sent a notification to its millions of users urging them to call their members of Congress to oppose the bill, Gallagher said it “proved” that the social media company had “brainwashed” American youth. While President Donald Trump temporarily reversed his Democratic predecessor’s TikTok ban the day after he took office, the application’s future in the country, as well as the state of its pro-Palestine content, remains unclear. When the application was available to Americans again in late January, many users pointed out that phrases like “free Palestine” were being flagged as hate speech, raising concerns about potential censorship on the platform following its return to the US.
Submission + - Argentinian president promotes $LIBRA cryptocoin... which crashes into oblivion
gwolf writes: On Friday, February 14, Libertarian Argentinian president, Javier Milei, promoted the just-created $LIBRA cryptocoin, created by the Viva la libertad project, strongly aligned with his political party, La Libertad Avanza. Milei tweeted, ÂThis private project will be devoted to promote growth of the Argentinian economy, funding small startups and enterprises. The world wants to invest in Argentina!Â. It is worth noting that the project's website was registered a mere three minutes before Milei tweeted his endorsement. The cryptocoin quickly reached a $4.6 billion market cap... Only to instantaneously lose 89% of its value, with nine core investers pulling the rug from under the enthusiast investers.
Of course, Milei angrily answered with a new tweet blaming everybody but himself. Is there any way to believe he wasn't aware of the shoddy associates he was promoting? Or that promoting a memecoin is not responsible for the head of state of a country?
Submission + - Are DOGE's Claims of Social Security Payments to 150-Year-Olds Way Off Base? 1
theodp writes: Fox News and other major news outlets reported that Elon Musk, who has been tasked with leading DOGE as a special government employee, spoke to reporters on Tuesday from the Oval Office of the White House with President Donald Trump and said DOGE found payments going to beneficiaries listed as being around the age of 150, though he didn't go into detail about the claims.
"There's crazy things, like, just a cursory examination of Social Security and we've got people in there that are about 150 years old," Musk said. "Now, do you know anyone that's 150? I don't. They should be in the Guinness Book of World Records, they're missing out. So, that's the case where, like, I think they're probably dead is my guess, or they should be very famous. One of the two," he added.
While BBC fact-checkers and the New York Times reported they could neither confirm nor deny Musk's claims, others on the web aren't buying Musk's story. Daily Kos contributor Lobachevsky offers this possible explanation for the 'crazy things' Musk credited his team for uncovering: "Reports say that his group at DOGE is made up of fairly young people. What those kids don’t realize is that Social Security uses VERY OLD computers. They’re programmed with an old version of the programming language COBOL. A bit of history. On May 20, 1875 a bunch of countries got together to create the International Bureau of Weight and Measures which established uniform standards of mass and length. Later on, the Bureau established rules for dates as well. The dates standard used a starting date of May 20 1875 to honor the creation of the Bureau. Old versions of COBOL use that date as a baseline. Social Security’s computers use that old version. Dates are stored as the number of days AFTER May 20 1875. So what happens if Social Security doesn’t know a birthdate? That field is empty in its records. Thus that person appears to have a birthday of May 20 1875-about 150 years ago. That’s why the crack team of youngsters Musk uses found 150-year-old people in Social Security getting benefits. It’s all really as simple—and as stupid—as that."
There are undoubtedly big fraud problems to solve at Social Security, just as there are at Musk-founded PayPal and other companies. But does Social Security truly have a material problem with paying people who are "about 150 years old," or is the problem here more one of misinformed sensationalism? And could Musk's DOGE team use a crash course in COBOL and other data representation arcana?
"There's crazy things, like, just a cursory examination of Social Security and we've got people in there that are about 150 years old," Musk said. "Now, do you know anyone that's 150? I don't. They should be in the Guinness Book of World Records, they're missing out. So, that's the case where, like, I think they're probably dead is my guess, or they should be very famous. One of the two," he added.
While BBC fact-checkers and the New York Times reported they could neither confirm nor deny Musk's claims, others on the web aren't buying Musk's story. Daily Kos contributor Lobachevsky offers this possible explanation for the 'crazy things' Musk credited his team for uncovering: "Reports say that his group at DOGE is made up of fairly young people. What those kids don’t realize is that Social Security uses VERY OLD computers. They’re programmed with an old version of the programming language COBOL. A bit of history. On May 20, 1875 a bunch of countries got together to create the International Bureau of Weight and Measures which established uniform standards of mass and length. Later on, the Bureau established rules for dates as well. The dates standard used a starting date of May 20 1875 to honor the creation of the Bureau. Old versions of COBOL use that date as a baseline. Social Security’s computers use that old version. Dates are stored as the number of days AFTER May 20 1875. So what happens if Social Security doesn’t know a birthdate? That field is empty in its records. Thus that person appears to have a birthday of May 20 1875-about 150 years ago. That’s why the crack team of youngsters Musk uses found 150-year-old people in Social Security getting benefits. It’s all really as simple—and as stupid—as that."
There are undoubtedly big fraud problems to solve at Social Security, just as there are at Musk-founded PayPal and other companies. But does Social Security truly have a material problem with paying people who are "about 150 years old," or is the problem here more one of misinformed sensationalism? And could Musk's DOGE team use a crash course in COBOL and other data representation arcana?
Submission + - The World's Most Printed 3D Model, 3DBenchy, is Now Open Source! (fabbaloo.com)
An anonymous reader writes: The original design was produced by Daniel Norée and Paulo Kiefe at Creative Tools in Sweden, who sought a way to promote their sales of 3D printers by establishing a fun benchmark. Their strategy worked, but Creative Tools eventually moved away from 3D printing and was later acquired by NTI Group, a much larger corporation.
Initially, it was thought that NTI Group had enforced their licensing rights on Printables, but it turned out not to be the case.
Now we see a resolution to this design mess.
In an announcement from NTI Group, they say this:
“3DBenchy, a 3D model designed specifically for testing and benchmarking 3D printers, is now in the public domain. Originally released on April 9, 2015, by Creative Tools, the model has become a beloved icon of the 3D printing community. Following Creative Tools’ acquisition and merger into NTI in 2024, NTI has decided to release 3DBenchy to the world by making it public domain, marking its 10th anniversary with this significant gesture.”
The company has also transferred control of the website 3Dbenchy.com to Daniel Norée and Paulo Kiefe. However, there is one stipulation: that the #3DBenchy “remains publicly available to all.”
Initially, it was thought that NTI Group had enforced their licensing rights on Printables, but it turned out not to be the case.
Now we see a resolution to this design mess.
In an announcement from NTI Group, they say this:
“3DBenchy, a 3D model designed specifically for testing and benchmarking 3D printers, is now in the public domain. Originally released on April 9, 2015, by Creative Tools, the model has become a beloved icon of the 3D printing community. Following Creative Tools’ acquisition and merger into NTI in 2024, NTI has decided to release 3DBenchy to the world by making it public domain, marking its 10th anniversary with this significant gesture.”
The company has also transferred control of the website 3Dbenchy.com to Daniel Norée and Paulo Kiefe. However, there is one stipulation: that the #3DBenchy “remains publicly available to all.”
Submission + - AskSlashdot: What would it require for you to trust an AI? (win.tue.nl) 2
shanen writes: Do you trust AI in general? What about DeepSeek in particular? Can anyone even explain what "trust" means these days? For example, I bet most of you trust Amazon and Amazon's secret AIs more than you should...
Recently I've been dabblling with a number of the GAI websites. Of course if I was a serious scientist then I'd have my own hardware and be running local tests. Or at least I'd be rooting around arXiv to see what the serious scientists are saying... All I have (as usual) are too many questions and too few jokes. Especially funny jokes. I wouldn't recognize a funny AI joke until after the AI had pwned me to pieces...
Actually the funniest joke I can think of on this topic involves DeepSeek. One of my experimental conversations on that website involved the topic of trust. DeepSeek turned out to be extremely good at explaining why I should not trust it. Every computer security problem I ever thought of or heard about and some more besides. "So what's that got to do with the price of tea in China?"
It's like the accountant who gets asked what 2 plus 2 is. After locking the doors and shading all the windows, the accountant whispers in your ear: "What do you want it to be?" And the price of tea in China is whatever Xi wants it to be! I bet you assumed the accountant was male, right? So the next questions are whether DeepSeek can do accounting or windows?
So let me start with some questions about DeepSeek in particular:
Have you run it locally and compared the responses with the website's responses? My hypothesis is that your mileage should differ...
It's well established that DeepSeek doesn't want to talk about many "political" topics. Is that based on a distorted model of the world? Or is the censorship implemented in the query interface after the model was trained? My hypothesis is that it must have been trained with lots of data because the cost of removing all of the bad stuff would have been prohibitive. Defining "bad" doesn't matter because I bet everyone agrees the Internet is chock full of bad data these years. Unless perhaps another AI filtered the data first?
What does trust mean? What sort of responses am I hoping to see? How many people still using today's Slashdot have even heard of "Reflections on Trusting Trust"? How many of the identities on today's Slashdot are just (AI-driven?) sock puppets? (Speculations on a mutual timeline building tool to verify childhood friendships?)
In closing, if you asked an AI to analyze all of the conversations on Slashdot, what sort of changes would it show over the years? My hypothesis on this question is that the interactions based on books will trend down. Maybe that's my selective memory, but I think a "sound" analysis would should a monotonic decrease. Largely based on The Anxious Generation by Jonathan Haidt I think the downward gradient might peak after smartphones became widely adopted circa 2010...
Recently I've been dabblling with a number of the GAI websites. Of course if I was a serious scientist then I'd have my own hardware and be running local tests. Or at least I'd be rooting around arXiv to see what the serious scientists are saying... All I have (as usual) are too many questions and too few jokes. Especially funny jokes. I wouldn't recognize a funny AI joke until after the AI had pwned me to pieces...
Actually the funniest joke I can think of on this topic involves DeepSeek. One of my experimental conversations on that website involved the topic of trust. DeepSeek turned out to be extremely good at explaining why I should not trust it. Every computer security problem I ever thought of or heard about and some more besides. "So what's that got to do with the price of tea in China?"
It's like the accountant who gets asked what 2 plus 2 is. After locking the doors and shading all the windows, the accountant whispers in your ear: "What do you want it to be?" And the price of tea in China is whatever Xi wants it to be! I bet you assumed the accountant was male, right? So the next questions are whether DeepSeek can do accounting or windows?
So let me start with some questions about DeepSeek in particular:
Have you run it locally and compared the responses with the website's responses? My hypothesis is that your mileage should differ...
It's well established that DeepSeek doesn't want to talk about many "political" topics. Is that based on a distorted model of the world? Or is the censorship implemented in the query interface after the model was trained? My hypothesis is that it must have been trained with lots of data because the cost of removing all of the bad stuff would have been prohibitive. Defining "bad" doesn't matter because I bet everyone agrees the Internet is chock full of bad data these years. Unless perhaps another AI filtered the data first?
What does trust mean? What sort of responses am I hoping to see? How many people still using today's Slashdot have even heard of "Reflections on Trusting Trust"? How many of the identities on today's Slashdot are just (AI-driven?) sock puppets? (Speculations on a mutual timeline building tool to verify childhood friendships?)
In closing, if you asked an AI to analyze all of the conversations on Slashdot, what sort of changes would it show over the years? My hypothesis on this question is that the interactions based on books will trend down. Maybe that's my selective memory, but I think a "sound" analysis would should a monotonic decrease. Largely based on The Anxious Generation by Jonathan Haidt I think the downward gradient might peak after smartphones became widely adopted circa 2010...
Submission + - Trump Admin to Purchase $400m in Armored EVs (npr.org)
Brentyl writes: According to this article from NPR, the Trump administration intends to buy $400 million worth of armored electric vehicles.
An early version of the official statement specified that the vehicles would be Teslas, until the statement was edited to remove the specific brand.
Critics point to the obvious apparent conflict of interest, with Elon Musk having a prominent role in the Trump administration.
An early version of the official statement specified that the vehicles would be Teslas, until the statement was edited to remove the specific brand.
Critics point to the obvious apparent conflict of interest, with Elon Musk having a prominent role in the Trump administration.
Comment Re:Mother Theresa is an unfortunate choice (Score 1) 273
But she did, in fact, preside over awful standards of care, people were denied access to medical treatment, and suffering was not alleviated, because it was considered "spiritually noble".
I have previously looked into those allegations. While she may have believed "suffering is good for the soul," it wasn't so much a denial of pain medication as a lack of access to them. Many of these clinics that were setup were in places where access to any medical care was absent.
Considering that pharmaceutical companies were lining up in droves to ship her free medications, and she refused, saying her vow of poverty prevented her from accepting free stuff for her patients, this "but there was a lack of access" argument is flimsy at best. She also refused free diagnosis charts, which could have helped prevent her nuns from misdiagnosing malaria in a young boy as a chest cold, as Dr. Robin Fox noted in his article in The Lancet in 1991. Refusing free stuff isn't a lack of access to stuff, since the access to that stuff is readily available and no one was going to charge money for the stuff.. Refusing stuff means Mother Theresa and her nuns were actively denying care to their patients.
Comment Re:Mother Theresa is an unfortunate choice (Score 3, Informative) 273
So you're saying she was a Republican?
Backwards. The party with a vested interest in keeping people dependent on professionals who dole things out to them is the Democrats. That's the backbone of their entire constituency and the framework within which they describe everybody: needing a handout, or needing to be used to pay for handouts. Without playing middlemen to that one-way street, there would be almost not power in that camp. And so they seek to preserve it at every turn.
No, the guy to whom you replied got it right: Republicans are the most dependent on a culture of people dependent on professionals who dole things out to them. Red States are more dependent on the Government Dole than Blue States, because Red State policies create a constituency which needs a handout just to survive. Poverty-stricken, uneducated white people vote Republican more often than middle class educated people (who tend to vote Democrat), so Republicans seek to preserve a constituency trapped in poverty, voting Republican on social issues even as Republicans pull the economic rug out from under their collective feet.
Slashdot Top Deals
All the evidence concerning the universe has not yet been collected, so there's still hope.
