Group items matching

in title, tags, annotations or url

""As the deceptive capabilities of AI systems become more advanced, the dangers they pose to society will become increasingly serious," said Dr Peter Park, an AI existential safety researcher at MIT and author of the research. Park was prompted to investigate after Meta, which owns Facebook, developed a program called Cicero that performed in the top 10% of human players at the world conquest strategy game Diplomacy. Meta stated that Cicero had been trained to be "largely honest and helpful" and to "never intentionally backstab" its human allies."

"If you were doubting how important recommender systems are to social media companies, a lawsuit filed last week against Meta makes it crystal clear. At the heart of this legal battle is a fundamental question: Shouldn't users have the power to decide what they do and don't see online? The lawsuit filed by Knight First Amendment Institute at Columbia University on behalf of Professor Ethan Zuckerman directly challenges how social media feeds are curated. Professor Zuckerman's proposed browser extension, 'Unfollow Everything 2.0,' would enable Facebook users to disengage from the algorithmically driven content that dominates their feeds, by allowing them to unfollow friends, pages and groups en masse, thus resetting their digital interactions on their terms."

""Elderly Chinese still often prefer to pay with cash and some struggle with using mobile payments." Less than a year ago, state media was lauding China's trajectory towards becoming the world's top country for cashless transactions. Xinhua reported cash had dropped to just 3.7% of the total money in circulation. But in recent months China's government has appeared to push back, with numerous announcements about "streamlining" payment systems for visitors and elderly people."

"Can AI ever be unbiased? As AI systems become more integrated into our daily lives, it's crucial that we understand the complexities of bias and how it impacts these technologies. From chatbots to hiring algorithms, the potential for AI to perpetuate and even amplify existing biases is a genuine concern. "

"The new AI Index report notes that in 2023, AI still struggled with complex cognitive tasks like advanced math problem-solving and visual commonsense reasoning. However, 'struggled' here might be misleading; it certainly doesn't mean AI did badly. Performance on MATH, a dataset of 12,500 challenging competition-level math problems, improved dramatically in the two years since its introduction. In 2021, AI systems could solve only 6.9% of problems. By contrast, in 2023, a GPT-4-based model solved 84.3%. The human baseline is 90%. "

"The Israeli military's bombing campaign in Gaza used a previously undisclosed AI-powered database that at one stage identified 37,000 potential targets based on their apparent links to Hamas, according to intelligence sources involved in the war. In addition to talking about their use of the AI system, called Lavender, the intelligence sources claim that Israeli military officials permitted large numbers of Palestinian civilians to be killed, particularly during the early weeks and months of the conflict."

"Much of west and central Africa has been left without internet service, as operators of several subsea cables reported failures. The cause of the cable failures on Thursday was not immediately clear. The African subsea cable operator Seacom confirmed that services on its west African cable system were down and that customers who relied on that cable were being redirected to the Google Equiano cable, which Seacom uses."

""This photo is a prime example of why 2024 is a crucial year for spotting - and stopping - manipulated media," says Shweta Singh, an assistant professor of information systems at Warwick Business School. "Whilst this may have been some low-level photoshopping, much of the edited media currently circulating can be more sinister. With elections in both the UK and the US this year, the importance of media being genuine has never been higher. Suspect photoshopping like this only undermines the faith of the public in the media they are presented with, and risks seriously damaging public trust.""

"Many researchers steeped in these issues, including myself, expect that the most likely result of building a superhumanly smart AI, under anything remotely like the current circumstances, is that literally everyone on Earth will die. Not as in "maybe possibly some remote chance," but as in "that is the obvious thing that would happen." It's not that you can't, in principle, survive creating something much smarter than you; it's that it would require precision and preparation and new scientific insights, and probably not having AI systems composed of giant inscrutable arrays of fractional numbers."

"The problem extends beyond the Pegasus project. Installed in Mexico City is one of the largest urban surveillance systems in the Americas: El Centro de Comando, Control, Cómputo, Comunicaciones y Contacto Ciudadano, better known as El C5. The network, connected to panic buttons and command centers, is spread over 1,485 kilometers with software designed to automatically detect license plates. On top of that, the number of installed cameras grew from 18 million to 65 million between 2018 and 2022, with stated plans to add at least an additional 16 million more. Despite its apparent pre-eminence, issues have arisen with the C5, from false identifications to mishandling of personal data. Technological malfunctions have also been shown to impact the outcomes of criminal cases because of the assumption of objectivity that video surveillance supposedly construes. The sprawling C5 system is dwarfed only by the Titan, an expansive intelligence and security database, both in terms of scale and threat to civil liberties. The software is used by several Mexican state governments to combine location data with other private information, including financial, government, and telecom data, to geolocate individuals across the country in real time. Governmental officials have been criticized for the controversial use of the database to target public figures, but, more problematically, access to Titan-enabled intel can be gained through an underground market, making it a further liability. The extent to which artificial intelligence has been incorporated into the C5 and Titan is still not clear, but the specter of surveillance remains large and is set to cause more worries with the addition of new smart technologies."

" A finance worker at a multinational firm was tricked into paying out $25 million to fraudsters using deepfake technology to pose as the company's chief financial officer in a video conference call, according to Hong Kong police. The elaborate scam saw the worker duped into attending a video call with what he thought were several other members of staff, but all of whom were in fact deepfake recreations, Hong Kong police said at a briefing on Friday. "(In the) multi-person video conference, it turns out that everyone [he saw] was fake," senior superintendent Baron Chan Shun-ching told the city's public broadcaster RTHK. Chan said the worker had grown suspicious after he received a message that was purportedly from the company's UK-based chief financial officer. Initially, the worker suspected it was a phishing email, as it talked of the need for a secret transaction to be carried out. However, the worker put aside his early doubts after the video call because other people in attendance had looked and sounded just like colleagues he recognized, Chan said."

"For Bejar, the controls in place on social networks like Instagram are not sufficient because they turn "inherently human interactions into an objective assessment". There are too few options for users to hide content or flag comments and DMs and explain why it made them uncomfortable even if it doesn't violate Meta's specific policies, he said. "There's a question of how clearly bad does the content need to be to warrant removal? And that means you set a line somewhere and have to define a criterion where either a computer system or a human can evaluate a piece of content," Bejar said."

"Performance of AMIE In this setting, we observed that AMIE performed simulated diagnostic conversations at least as well as PCPs when both were evaluated along multiple clinically-meaningful axes of consultation quality. AMIE had greater diagnostic accuracy and superior performance for 28 of 32 axes from the perspective of specialist physicians, and 24 of 26 axes from the perspective of patient actors. "

"Next, was two months of probation where we moderated on practice queues that consisted of hundreds of thousands of videos that had already been moderated. The policies we applied to these practice videos were compared with what had previously been applied to them by a more experienced moderator in order to find areas we needed to improve in. Everyone passed their probation. One trend that is particularly hated by moderators are the "recaps". These consist of a 15- to 60-second barrage of pictures, sometimes hundreds, shown as a super fast slideshow often with three to four pictures a second. We have to view every one of these photos for infractions. If a video is 60 seconds long then the system will allocate us around 48 seconds to do this. We also have to check the video description, account bio and hashtags. Around the end of the school year or New Year's Eve, when these sort of videos are popular, it becomes incredibly draining and also affects our stats. "

"The UK is not Russia. For all that the many civil liberty campaigners will complain, as is their role, the independence of the judiciary remains strong. The laws relating to freedom of association, expression and right to privacy are well defended in parliament and outside. But the technology, the means by which the state might insert itself into our lives, is developing apace. The checks and balances are not. The Guardian has revealed that the government is legislating, without fanfare, to allow the police and the National Crime Agency to run facial recognition searches across the UK's driving licence records. When the police have an image, they will be able to identify the person, it is hoped, through the photographic images the state holds for the purposes of ensuring that the roads are safe. Searching those digital images would have taken more man-hours than could have been justified in the old analogue world. It is now a matter of pushing a button, thanks to the wonders of artificial intelligence systems that are able to match biometric measurements in a flash."

"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."

"Sources familiar with how AI-based systems have been integrated into the IDF's operations said such tools had significantly sped up the target creation process. "We prepare the targets automatically and work according to a checklist," a source who previously worked in the target division told +972/Local Call. "It really is like a factory. We work quickly and there is no time to delve deep into the target. The view is that we are judged according to how many targets we manage to generate." A separate source told the publication the Gospel had allowed the IDF to run a "mass assassination factory" in which the "emphasis is on quantity and not on quality". A human eye, they said, "will go over the targets before each attack, but it need not spend a lot of time on them". For some experts who research AI and international humanitarian law, an acceleration of this kind raises a number of concerns."

"Imagine having a digital twin that gets ill, and can be experimented on to identify the best possible treatment, without you having to go near a pill or a surgeon's knife. Scientists believe that within five to 10 years, "in silico" trials - in which hundreds of virtual organs are used to assess the safety and efficacy of drugs - could become routine, while patient-specific organ models could be used to personalise treatment and avoid medical complications. Digital twins are computational models of physical objects or processes, updated using data from their real-world counterparts. Within medicine, this means combining vast amounts of data about the workings of genes, proteins, cells and whole-body systems with patients' personal data to create virtual models of their organs - and eventually, potentially their entire body"

"The use for humans in the world of robotics, even as teachers, is shrinking thanks to AI. NVIDIA Research has announced the creation of Eureka, an AI agent powered by GPT-4 that has trained robots to perform tasks using reward algorithms. Notably, Eureka taught a robotic hand to do pen spinning tricks as well as a human can (honestly, as you can see in the YouTube video below, better than many of us)."

"Tesla denied liability, saying Lee consumed alcohol before getting behind the wheel. The electric-vehicle maker also claims it was unclear whether the autopilot feature was engaged at the time of the crash. Tesla has been testing and rolling out its autopilot and more advanced full self-driving (FSD) system, which its chief executive, Elon Musk, has touted as crucial to his company's future but has drawn regulatory and legal scrutiny. Tesla won an earlier trial in Los Angeles in April with a strategy of saying that it tells drivers that its technology requires human monitoring, despite the "autopilot" and "full self-driving" names."