Computers & Technology 4

Author	Message
ML1 Volunteer moderator Volunteer tester Send message Joined: 25 Nov 01 Posts: 20991 Credit: 7,508,002 RAC: 20	Message 2114714 - Posted: 14 Feb 2023, 23:12:42 UTC - in response to Message 2114450. Last modified: 14 Feb 2023, 23:14:25 UTC Here's a very good explanation of the "what", the 'fun', and the 'furor' surrounding, the Bard and ChatGPT "AI chatbots": ChatGPT Is a Blurry JPEG of the Web wrote: OpenAI’s chatbot offers paraphrases, whereas Google offers quotes. Which do we prefer? ... This analogy to lossy compression is not just a way to understand ChatGPT’s facility at repackaging information found on the Web by using different words. It’s also a way to understand the “hallucinations,” or nonsensical answers to factual questions, to which large language models such as ChatGPT are all too prone. These hallucinations are compression artifacts... ... The Web certainly contains explanations ... but GPT-3 isn’t able to incorporate those explanations. GPT-3’s statistical analysis of examples ... enables it to produce a superficial approximation of the real thing, but no more than that... ... Given GPT-3’s failure at a subject taught in elementary school, how can we explain the fact that it sometimes appears to perform well at writing college-level essays? Even though large language models often hallucinate, when they’re lucid they sound like they actually understand subjects... ... Is it possible that, in [some] areas ..., statistical regularities in text actually do correspond to genuine knowledge of the real world?... ... Can large language models take the place of traditional [internet] search engines?... That article itself is a beautiful example of exactly what those supposed "AI" chatbots simply cannot do at present. Roll onwards the development of the GPT-4 model?... Keep searchin', Martin See new freedom: Mageia Linux Take a look for yourself: Linux Format The Future is what We all make IT (GPLv3) ID: 2114714 · Reply Quote

W-K 666 Volunteer tester Send message Joined: 18 May 99 Posts: 19313 Credit: 40,757,560 RAC: 67	Message 2114724 - Posted: 15 Feb 2023, 2:17:21 UTC - in response to Message 2114714. Here's a completely different view, Tom Scott - I tried using AI. It scared me. ID: 2114724 · Reply Quote

Dr Who Fan Volunteer tester Send message Joined: 8 Jan 01 Posts: 3316 Credit: 715,342 RAC: 4	Message 2115248 - Posted: 25 Feb 2023, 21:58:21 UTC The Computer That Will Change Everything Eight years in the making, Aurora, a powerful new machine at Argonne National Laboratory, could help solve some of the most pressing questions of our time. Welcome to the new era of supercomputing. ID: 2115248 · Reply Quote

Dr Who Fan Volunteer tester Send message Joined: 8 Jan 01 Posts: 3316 Credit: 715,342 RAC: 4	Message 2115526 - Posted: 4 Mar 2023, 4:20:04 UTC The inside story of how ChatGPT was built from the people who made it. Exclusive conversations that take us behind the scenes of a cultural phenomenon. To get the inside story behind the chatbot—how it was made, how OpenAI has been updating it since release, and how its makers feel about its success—I talked to four people who helped build what has become one of the most popular internet apps ever. In addition to Agarwal and Fedus, I spoke to John Schulman, a cofounder of OpenAI, and Jan Leike, the leader of OpenAI’s alignment team, which works on the problem of making AI do what its users want it to do (and nothing more). ID: 2115526 · Reply Quote

Wiggo Send message Joined: 24 Jan 00 Posts: 36363 Credit: 261,360,520 RAC: 489	Message 2115694 - Posted: 7 Mar 2023, 11:01:41 UTC High Court dumps Facebook's challenge against prosecution over Cambridge Analytica privacy breaches. The High Court has refused to hear a challenge from Facebook, which argued it should not face prosecution for an alleged privacy breach that may have affected hundreds of thousands of Australians. The social media giant's lawyers appeared before the court this morning to argue it was not responsible for the impact of the Cambridge Analytica scandal on its Australian users. Australia's privacy watchdog is seeking to prosecute Facebook for providing allegedly unlawful access to the personal data of almost 90 milllion Facebook users worldwide — including some Australians — almost a decade ago. In 2014, British political consulting firm Cambridge Analytica helped develop software that went on to collect data from Facebook users without their knowledge or consent. The software — a personality quiz — not only pull information from those who used it, but also gathered data from anyone who was connected to those users on Facebook. In Australia, 53 people logged in to the online quiz, which is estimated to have exposed the data of up to 300,000 other Australians. Cambridge Analytica later used this vast database to create psychological profiles and to target political messages during Donald Trump's 2016 United States presidential election campaign. The Australian Information Commissioner (AIC) alleges that Facebook breached Australian privacy laws between March 12, 2014 and May 1, 2015. Facebook had applied to the High Court to dismiss that case, arguing it should not face prosecution in this country because, at that time, it did not conduct a business within Australia, and had no personnel or premises. That changed in 2018 — well after the alleged breaches — when the business established an Australian presence.... ID: 2115694 · Reply Quote

Dr Who Fan Volunteer tester Send message Joined: 8 Jan 01 Posts: 3316 Credit: 715,342 RAC: 4	Message 2116094 - Posted: 14 Mar 2023, 7:50:53 UTC Robot lawyer' that never got court day sued for practicing without license A company that was foiled by "threats from State Bar prosecutors" for its plans to use an artificial intelligence chatbot to advise a defendant in traffic court earlier this year is being sued by a top class-action firm that says the company is practicing law without a license. Chicago-based law firm Edelson wrote in a proposed class action that the San Francisco-based DoNotPay is "not actually a robot, a lawyer, nor a law firm," according to a 12-page March 3 filing in the San Francisco County Superior Court . ID: 2116094 · Reply Quote

Richard Haselgrove Volunteer tester Send message Joined: 4 Jul 99 Posts: 14674 Credit: 200,643,578 RAC: 874	Message 2116105 - Posted: 14 Mar 2023, 14:36:10 UTC I'm struggling to reconcile these two news stories. Tiny data centre used to heat public swimming pool Rishi Sunak has electricity grid upgraded to heat his private pool How can "a washing-machine-sized data centre" handle £100,000 worth of electricity, even at current prices? Or do a job that requires "a substantial amount of equipment and a new connection to the National Grid"? Given the high thermal capacity of water, what temperature does that washing machine have to run at? Or has someone found a way of breaking the laws of thermodynamics? ID: 2116105 · Reply Quote

ML1 Volunteer moderator Volunteer tester Send message Joined: 25 Nov 01 Posts: 20991 Credit: 7,508,002 RAC: 20	Message 2116174 - Posted: 15 Mar 2023, 23:04:26 UTC - in response to Message 2116105. ... How can "a washing-machine-sized data centre" handle £100,000 worth of electricity, even at current prices? Or do a job that requires "a substantial amount of equipment and a new connection to the National Grid"?... There's a good informative take on that story over on The Register: Cloud upstart offers free heat if you host its edge servers wrote: ... The technology has already been deployed at Exmouth Leisure Centre in south west England, where the waste heat from a dozen servers is being used to warm the swimming pool. This is expected to reduce the pool's energy requirements by 62 percent, saving them over £20,000 ($24,334) a year and also reducing their carbon dioxide emissions, we're told. This deployment will be followed by further installations in Bristol and Manchester in the near future, according to Deep Green. The company installs the kit for free, and also covers the electricity and maintenance costs of the infrastructure. Similar projects using waste heat from datacenters are already in use at various locations, though usually on a much larger scale... Very good and useful if you have a secure business location and the necessary electricity supply. Happy cool crunchin'! Martin See new freedom: Mageia Linux Take a look for yourself: Linux Format The Future is what We all make IT (GPLv3) ID: 2116174 · Reply Quote

ML1 Volunteer moderator Volunteer tester Send message Joined: 25 Nov 01 Posts: 20991 Credit: 7,508,002 RAC: 20	Message 2116217 - Posted: 16 Mar 2023, 20:35:45 UTC Ouch! YouTube - Would You Let AI Represent You In Court? Scarily all too real!! Keep ahead there! Martin See new freedom: Mageia Linux Take a look for yourself: Linux Format The Future is what We all make IT (GPLv3) ID: 2116217 · Reply Quote

Dr Who Fan Volunteer tester Send message Joined: 8 Jan 01 Posts: 3316 Credit: 715,342 RAC: 4	Message 2116233 - Posted: 16 Mar 2023, 23:57:15 UTC - in response to Message 2116217. Ouch! YouTube - Would You Let AI Represent You In Court? Scarily all too real!! Keep ahead there! Martin Oh he77 NO! Sounds just like the one from the article I posted several messages back. ID: 2116233 · Reply Quote

ML1 Volunteer moderator Volunteer tester Send message Joined: 25 Nov 01 Posts: 20991 Credit: 7,508,002 RAC: 20	Message 2116322 - Posted: 18 Mar 2023, 21:30:26 UTC No surprise... HP outrages printer users with firmware update suddenly bricking third-party ink wrote: ... "HP have updated their printers to outright ban ‘non-HP’ ink! They no longer shows the 'can’t guarantee quality' message, but instead cancels your print completely until you inset a HP ink cartridge,"... ... In fact, even when agreeing to pay users in class-action lawsuits, it has been careful not to admit wrongdoing... I wonder... Is that why I do not recommend any "ink-jet" printers, and why simply I just do not touch the things! IT is what we allow it to be... Martin See new freedom: Mageia Linux Take a look for yourself: Linux Format The Future is what We all make IT (GPLv3) ID: 2116322 · Reply Quote

Gary Charpentier Volunteer tester Send message Joined: 25 Dec 00 Posts: 30929 Credit: 53,134,872 RAC: 32	Message 2116371 - Posted: 19 Mar 2023, 15:34:49 UTC - in response to Message 2116322. Considering what has been done with USB, would you want your IoT, Printer, etc. to have unprotected cybersex with just any random chip? IT is what we allow it to be... ID: 2116371 · Reply Quote

ML1 Volunteer moderator Volunteer tester Send message Joined: 25 Nov 01 Posts: 20991 Credit: 7,508,002 RAC: 20	Message 2116386 - Posted: 19 Mar 2023, 20:17:29 UTC - in response to Message 2116371. Might that be why foreign USB devices are supposedly banned in all tech-savvy workplaces? ... And why sometimes all the USB ports on equipment/PCs are disabled with epoxy resin?... Except... How many people know the unfixable security perils of USB?... Is that why the USA and UK nuclear deterrents are protected by 8" floppy disks?... Stay safe folks!!! Martin See new freedom: Mageia Linux Take a look for yourself: Linux Format The Future is what We all make IT (GPLv3) ID: 2116386 · Reply Quote

ML1 Volunteer moderator Volunteer tester Send message Joined: 25 Nov 01 Posts: 20991 Credit: 7,508,002 RAC: 20	Message 2116387 - Posted: 19 Mar 2023, 20:21:41 UTC Is this one acronym too far for a better user experience?... Techie fired for inventing an acronym... wrote: A tale of how a PEBCAK became a CLE... ... It wasn't a happy job. "People really did not like tech support because they were a bit snobbish and off-putting." Well, that'll do it. When Hal turned up to solve users' problems he tried to change their attitude by cheerfully telling them "Don't worry about it, just..." ... users seemed to like it... Yep... I've seen that for real. Including the management over-aggrandizement incompetence in all their ignorance... IT is what we allow it to be... Martin See new freedom: Mageia Linux Take a look for yourself: Linux Format The Future is what We all make IT (GPLv3) ID: 2116387 · Reply Quote

ML1 Volunteer moderator Volunteer tester Send message Joined: 25 Nov 01 Posts: 20991 Credit: 7,508,002 RAC: 20	Message 2116393 - Posted: 19 Mar 2023, 20:43:28 UTC Wow!... Quite a turn around... UNIX co-creator Ken Thompson is a… what user now? wrote: Elder statesman of system software makes a shocking revelation... ... back from the pandemic with a bang: Ken Thompson as keynote speaker. In the Q&A at the end of his talk, Thompson made a surprising confession. Bell Labs researcher Ken Thompson was one of the developers of MULTICS, the ancestor of and inspiration for UNIX... [and LOTs more...]... ... Living programmers don't get much more eminent than this. He is a genius, who devised and built tools that have deeply affected millions of people... ... The real surprise is during the question-and-answer section at the end, though (at the 57¾ minute mark.) An audience member asked: "What's your operating system of choice today?"... Do not underestimate the significance. IT was what he made it... Martin See new freedom: Mageia Linux Take a look for yourself: Linux Format The Future is what We all make IT (GPLv3) ID: 2116393 · Reply Quote

Dr Who Fan Volunteer tester Send message Joined: 8 Jan 01 Posts: 3316 Credit: 715,342 RAC: 4	Message 2116878 - Posted: 30 Mar 2023, 20:35:12 UTC Elon Musk Signs Open Letter Urging AI Labs to Pump the Brakes An open letter with signatures from hundreds of the biggest names in tech, including Elon Musk, has urged the world’s leading artificial intelligence labs to pause the training of new super-powerful systems for six months, saying that recent advances in AI present “profound risks to society and humanity.” The letter comes just two weeks after the public release of OpenAI’s GPT-4, the most powerful AI system ever released, which has led researchers to slash their expectations for when AGI—or artificial general intelligence that surpasses human cognitive ability—will arrive. Many experts fear that, as an AI arms race heats up, humanity is sleepwalking into catastrophe. “Advanced AI could represent a profound change in the history of life on Earth, and should be planned for and managed with commensurate care and resources,” the letter says. “Unfortunately, this level of planning and management is not happening, even though recent months have seen AI labs locked in an out-of-control race to develop and deploy ever more powerful digital minds that no one – not even their creators – can understand, predict, or reliably control.” ID: 2116878 · Reply Quote

Dr Who Fan Volunteer tester Send message Joined: 8 Jan 01 Posts: 3316 Credit: 715,342 RAC: 4	Message 2116919 - Posted: 31 Mar 2023, 19:02:09 UTC Italy Blocks Artificial Intelligence Bot ChatGPT over Privacy Concerns Italy’s Guarantor for the Protection of Personal Data announced (linked documents in Italian) on Friday that OpenAI, the Silicon Valley tech company behind the ChatGPT software application used to simulate and process human-like conversations, will be temporarily banned from accessing Italian user data through which it develops its algorithms and machine learning until it complies with privacy regulations. The regulator went on to say that it has opened up an investigation in response to an alleged data breach “regarding user conversations and information related to the payment of subscribers to the paid service.” ID: 2116919 · Reply Quote

ML1 Volunteer moderator Volunteer tester Send message Joined: 25 Nov 01 Posts: 20991 Credit: 7,508,002 RAC: 20	Message 2117298 - Posted: 8 Apr 2023, 19:29:56 UTC This just had to happen... ChatGPT: Mayor starts legal bid over false bribery claim wrote: An Australian mayor said he may take legal action over false information shared by advanced chatbot ChatGPT... ... His lawyers have sent a concerns notice to OpenAI - the first formal step in defamation action in Australia. OpenAI has 28 days to respond to the concerns notice, after which time Mr Hood would be able to take the company to court under Australian law. If he pursues the legal claim, it would be the first time OpenAI has publicly faced a defamation suit over the content created by ChatGPT... ... 'Plausible-sounding but incorrect' When people use ChatGPT, they are shown a disclaimer warning that the content it generates may contain "inaccurate information about people, places, or facts". And on its public blog about the tool, OpenAI says a limitation is that it "sometimes writes plausible-sounding but incorrect or nonsensical answers"... Do we not have enough Fake News to not have Chatbots making more of it up? Oh wait... Is Foxy Entertainment in truth merely a money spinning Murdoch Chatbot?... Enjoy?! Martin See new freedom: Mageia Linux Take a look for yourself: Linux Format The Future is what We all make IT (GPLv3) ID: 2117298 · Reply Quote

ML1 Volunteer moderator Volunteer tester Send message Joined: 25 Nov 01 Posts: 20991 Credit: 7,508,002 RAC: 20	Message 2117470 - Posted: 11 Apr 2023, 22:53:50 UTC Last modified: 11 Apr 2023, 22:54:12 UTC Quote of the day: Phoronix discussion comment wrote: Yes. The power of the default is a phenomenon that can never be overestimated. The only logical way out is to change the system: Linux on the desktop. Beware what you wish upon?! (Aside: Doesn't Microsoft now bundle Linux in with their Windows?...) Enjoy? Martin See new freedom: Mageia Linux Take a look for yourself: Linux Format The Future is what We all make IT (GPLv3) ID: 2117470 · Reply Quote

ML1 Volunteer moderator Volunteer tester Send message Joined: 25 Nov 01 Posts: 20991 Credit: 7,508,002 RAC: 20	Message 2117509 - Posted: 12 Apr 2023, 15:21:00 UTC Last modified: 12 Apr 2023, 15:21:20 UTC Twittered? Elon Musk: What it's like to interview the billionaire Twitter boss wrote: ... On Tuesday lunchtime I typed an email, requesting a chat about his first six months as Twitter boss. To my surprise, he responded. "Sure, how about tonight?" he said. I wasn't totally sure if he was being serious... So: Spontaneous? Brash? Rash? Eccentrically brilliant? Or all just part of being enterprising? See for yourselves! Martin See new freedom: Mageia Linux Take a look for yourself: Linux Format The Future is what We all make IT (GPLv3) ID: 2117509 · Reply Quote

©2024 University of California

SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.