Group items tagged

"In October, American teachers prevailed in a lawsuit with their school district over a computer program that assessed their performance. The system rated teachers in Houston by comparing their students' test scores against state averages. Those with high ratings won praise and even bonuses. Those who fared poorly faced the sack. The program did not please everyone. Some teachers felt that the system marked them down without good reason. But they had no way of checking if the program was fair or faulty: the company that built the software, the SAS Institute, regards its algorithm a trade secret and would not disclose its workings."

"In a paper titled "Towards Causal Representation Learning," researchers at the Max Planck Institute for Intelligent Systems, the Montreal Institute for Learning Algorithms (Mila), and Google Research discuss the challenges arising from the lack of causal representations in machine learning models and provide directions for creating artificial intelligence systems that can learn causal representations."

"SoundCloud announced on Tuesday it would become the first major streaming service to start directing subscribers' fees only to the artists they listen to, a move welcomed by musicians campaigning for fairer pay. Current practice for streaming services including Spotify, Deezer and Apple is to pool royalty payments and dish them out based on which artists have the most global plays. Many artists and unions have criticised this system, saying it disproportionately favours megastars and leaves y little for musicians further down the pecking order."

"Oliver also pointed to a Finnish man who once found "one of the most severe security flaws ever discovered in a voting system" in US machines and alerted their manufacturers, who released a patch to fix the problem in 2006. The state of Georgia, however, never installed it, and the Senate report noted their machines hadn't been updated since at least 2005. "They'd essentially been hitting the 'remind me tomorrow' button on a critical security update for over a decade," Oliver explained, "meaning Georgia's election systems operate on the same level of technical proficiency as Every Dad"."

"A flagship artificial intelligence system designed to predict gun and knife violence before it happens had serious flaws that made it unusable, police have admitted. The error led to large drops in accuracy and the system was ultimately rejected by all of the experts reviewing it for ethical problems."

"This week, the company released a new form of facial recognition called periocular recognition, which can supposedly identify individuals by just their eyes and eyebrows. Rank One says the new system uses an entirely different algorithm from its standard facial recognition system and is specifically meant for masked individuals. Rank One says it will ship the technology to all of its active customers for free."

"They then analysed how diverse the individuals' work was at different points in their careers. This was assessed using an artificial intelligence system that was trained, in the case of art, to "recognise" different styles by features such as the brush strokes, shapes and objects in a piece, while in the case of film, it was trained to classify a director's work based on plot and cast information. For science, the system identified different research topics based on the papers cited within a researcher's publications."

"Social media giant, Facebook, announced Tuesday that it is shutting down its facial recognition system which automatically identifies users in photos and videos, citing growing societal concerns about the use of such technology. "Regulators are still in the process of providing a clear set of rules governing its use," said Jerome Pesenti, vice president of artificial intelligence at Facebook, in a blog post. "Amid this ongoing uncertainty, we believe that limiting the use of facial recognition to a narrow set of use cases is appropriate.""

"The role of AI writing systems as drafting buddies is a big departure from how writers typically get help, yet so far it is their biggest selling point and use case. Most writing tools available today will do some drafting for you, either by continuing where you left off or responding to a more specific instruction. SudoWrite, a popular AI writing tool for novelists, does all of these, with options to "write" where you left off, "describe" a highlighted noun, or "brainstorm" ideas based on a situation you describe. Systems like Jasper.ai or Lex will complete your paragraph or draft copy based on instructions, and Laika is similar but more focused on fiction and drama. "

"According to these internal documents, SIAM is a computer system that works behind the scenes of Iranian cellular networks, providing its operators a broad menu of remote commands to alter, disrupt, and monitor how customers use their phones. The tools can slow their data connections to a crawl, break the encryption of phone calls, track the movements of individuals or large groups, and produce detailed metadata summaries of who spoke to whom, when, and where. Such a system could help the government invisibly quash the ongoing protests - or those of tomorrow - an expert who reviewed the SIAM documents told The Intercept."

"As Tremblay notes, a major problem for AI chatbots like ChatGPT and Bard is their tendency to confidently state incorrect information as fact. The systems frequently "hallucinate" - that is, make up information - because they are essentially autocomplete systems."

"They still have far more advanced technology that they haven't made publicly available yet. Something that does more or less what Bard does could have been released over two years ago. They've had that technology for over two years. What they've spent the intervening two years doing is working on the safety of it - making sure that it doesn't make things up too often, making sure that it doesn't have racial or gender biases, or political biases, things like that. That's what they spent those two years doing. But the basic existence of that technology is years old, at this point. And in those two years, it wasn't like they weren't inventing other things. There are plenty of other systems that give Google's AI more capabilities, more features, make it smarter. The most sophisticated system I ever got to play with was heavily multimodal - not just incorporating images, but incorporating sounds, giving it access to the Google Books API, giving it access to essentially every API backend that Google had, and allowing it to just gain an understanding of all of it."

"Sam Smith, of the health data privacy group MedConfidential, said: "This is an utterly appalling case. It's an individual problem that the doctor did this. But it's a systemic problem that they could do it, and that flaws in the way the NHS's data management systems work meant that any doctor can do something like this to any patient."

"Rite Aid facial recognition misidentified Black, Latino and Asian people as 'likely' shoplifters Surveillance systems incorrectly and without customer consent marked shoppers as 'persons of interest', an FTC settlement says Johana Bhuiyan and agencies Wed 20 Dec 2023 14.29 EST Last modified on Thu 21 Dec 2023 12.04 EST Rite Aid used facial recognition systems to identify shoppers that were previously deemed "likely to engage" in shoplifting without customer consent and misidentified people - particularly women and Black, Latino or Asian people - on "numerous" occasions, according to a new settlement with the Federal Trade Commission. As part of the settlement, Rite Aid has been forbidden from deploying facial recognition technology in its stores for five years."

"The university presidents' council said this morning that the security of its central admissions system has been upgraded after the personal data of around 23,000 students was advertised for sale on the dark web. The university presidents' council insisted that the leaked admissions data, which included personal information and examination results, was only current through last May after anger erupted over the data breach. #BanTCAS was surging on Thai Twitter this morning in reference to the Thai University Central Admission System, or TCAS."

"Kellin Pelrine, an American player who is one level below the top amateur ranking, beat the machine by taking advantage of a previously unknown flaw that had been identified by another computer. But the head-to-head confrontation in which he won 14 of 15 games was undertaken without direct computer support. The triumph, which has not previously been reported, highlighted a weakness in the best Go computer programs that is shared by most of today's widely used AI systems, including the ChatGPT chatbot created by San Francisco-based OpenAI. The tactics that put a human back on top on the Go board were suggested by a computer program that had probed the AI systems looking for weaknesses. The suggested plan was then ruthlessly delivered by Pelrine."

"The technology certainly has its flaws. While the system is theoretically designed not to cross some moral red lines - it's adamant that Hitler was bad - it's not difficult to trick the AI into sharing advice on how to engage in all sorts of evil and nefarious activities, particularly if you tell the chatbot that it's writing fiction. The system, like other AI models, can also say biased and offensive things. As my colleague Sigal Samuel has explained, an earlier version of GPT generated extremely Islamophobic content, and also produced some pretty concerning talking points about the treatment of Uyghur Muslims in China."

"Applications of these systems have been plagued by persistent inaccuracies in their output; these are often called "AI hallucinations". We argue that these falsehoods, and the overall activity of large language models, is better understood as bullshit in the sense explored by Frankfurt (On Bullshit, Princeton, 2005): the models are in an important way indifferent to the truth of their outputs. We distinguish two ways in which the models can be said to be bullshitters, and argue that they clearly meet at least one of these definitions. We further argue that describing AI misrepresentations as bullshit is both a more useful and more accurate way of predicting and discussing the behaviour of these systems."

"In a sharply worded warning, the cancer experts say that 'novel solutions' such as new diagnostic tests have been wrongly hyped as 'magic bullets' for the cancer crisis, but 'none address the fundamental issues of cancer as a systems problem'. A 'common fallacy' of NHS leaders is the assumption that new technologies can reverse inequalities, the authors add. The reality is that tools such as AI can create 'additional barriers for those with poor digital or health literacy'. 'We caution against technocentric approaches without robust evaluation from an equity perspective,' the paper concludes."

"The European Commission has asked X to hand over internal documents about its algorithms, as it steps up its investigation into whether Elon Musk's social media platform has breached EU rules on content moderation. The EU's executive branch told the company it wanted to see internal documentation about its "recommender system", which makes content suggestions to users, and any recent changes made to it, by 15 February. X has been under investigation since December 2023 under the EU's content law - known as the Digital Services Act (DSA) - over how it tackles the spread of illegal content and information manipulation. The company has been accused of manipulating the platform's systems to give far-right posts and politicians greater visibility over other political groups."