Updated Facial recognition technology is being employed in more UK schools to allow pupils to pay for their meals, according to reports today. In North Ayrshire Council, a Scottish authority encompassing the Isle of Arran, nine schools are set to begin processing meal payments for school lunches using facial scanning technology. The authority and the company implementing the technology, CRB Cunninghams, claim the system will help reduce queues and is less likely to spread COVID-19 than card payments and fingerprint scanners, according to the Financial Times. Speaking to the publication, David Swanston, the MD of supplier CRB Cunninghams, said the cameras verify the child's identity against "encrypted faceprint templates", and will be held on servers on-site at the 65 schools that have so far signed up. He added: "In a secondary school you have around about a 25-minute period to serve potentially 1,000 pupils. So we need fast throughput at the point of sale."

Naturally, we're biased to believe that we're the best -- just like you -- but we think it's important to question things sometimes. Technology can be biased -- even if it's not its fault. There are plenty of reasons to be wary about artificial intelligence. AI systems can inherit biases from their creators and from their surroundings. If the goal is to eliminate bias, it's important to consider AI bias. As researchers found, Google's image recognition algorithms labelled people of color as gorillas or that Microsoft's AI chatbot Tay rapidly became racist and sexist, and these biases were reinforced by how it was taught.

Policymakers around the world are developing guidelines for use of artificial intelligence in health care. Baymax, the robotic health aide and unlikely hero from the movie Big Hero 6, is an adorable cartoon character, an outlandish vision of a high-tech future. But underlying Baymax's character is the very realistic concept of an artificial intelligence (AI) system that can be applied to health care. As AI technology advances, how will regulators encourage innovation while protecting patient safety? AI does not have a precise definition, but the term generally describes machines that have the capacity to process and respond to stimulation in a manner similar to human thought processes.

Apple's second fall product event kicks off later today at 1 PM ET. We've laid out what to expect, but it's not the only big tech event week. Spare a thought for some of our staff, who will go straight from Apple reportage into Google. Yep, Tuesday October 19th is Google's Pixel 6 event. While we know what the phone will look like -- and some of its specifications -- expect to see some software surprises.

In the fall of 2015, Rob Reich, a philosopher and a political scientist at Stanford, was chatting with a freshman during office hours. "I asked him what he planned to study," Reich recalled recently. "He said, 'Definitely computer science. I have some ideas for startups.' " In the spirit of small talk, Reich asked, What kind? "He looked at me with total earnestness and said, 'To tell you that, I'd have to ask you to sign a nondisclosure agreement.'

Schools in Scotland are trialling facial recognition to allow pupils to pay for their lunches from Monday. The software is to be trialled across nine schools in North Ayrhsire, and hopes to speed up lunchtime sales by scanning the faces of pupils when at tills. Many schools already use biometric software, such as fingerprint recognition, to take payments but facial recognition is billed as being quicker and more Covid-secure. David Swanston, the manging director of CBR Cunninghams, who installed the software, said it was "the fastest way of recognising someone at the till." "In a secondary school you have around about a 25-minute period to serve potentially 1,000 pupils. So we need fast throughput at the point of sale, he told the Financial Times. MPs and peers urge education secretary to rethink plans to scrap most BTECs Covid cases near peak of second wave as schoolchildren fuel rise Schools reminded to allow absences arising from Covid'in exceptional circumstances' Schools reminded to allow absences arising from Covid'in exceptional circumstances' Mr Swanston said the software cut the average transaction time five seconds per pupil. But the new system has been criticised by privacy campaigners who say it normalises facial recognition software where there is little need for and was often operated without clear consent from the user. Silkie Carlo, of the civil liberties campaign group Big Brother Watch, said: "It's normalising biometric identity checks for something that is mundane.

The Government of India is bringing in reforms to allow private enterprises to participate in end-to-end space activities and help achieve the country's goal of enhancing its share in the global economy. The government officials also highlight the importance of space robotics and artificial intelligence in the space domain. "India is revising its existing policies and is also in the process of bringing in new ones to increase industry participation in the space sector," Indian Space Research Organisation (ISRO) Chairman Dr K. Sivan said, via a video message during the inaugural session on'Future of Space-International Participation and Collaborations' at The India Pavilion, Expo 2020 Dubai. He emphasized that the recent reforms in the sector has ensured that the role of the private sector has evolved from being just suppliers to partners in the process. He also highlighted that space is one of the significant areas that India is looking at for international cooperation.

What happens when a machine learning dataset is deprecated for legal, ethical, or technical reasons, but continues to be widely used? In this paper, we examine the public afterlives of several prominent deprecated or redacted datasets, including ImageNet, 80 Million Tiny Images, MS-Celeb-1M, Duke MTMC, Brainwash, and HRT Transgender, in order to inform a framework for more consistent, ethical, and accountable dataset deprecation. Building on prior research, we find that there is a lack of consistency, transparency, and centralized sourcing of information on the deprecation of datasets, and as such, these datasets and their derivatives continue to be cited in papers and circulate online. These datasets that never die -- which we term "zombie datasets" -- continue to inform the design of production-level systems, causing technical, legal, and ethical challenges; in so doing, they risk perpetuating the harms that prompted their supposed withdrawal, including concerns around bias, discrimination, and privacy. Based on this analysis, we propose a Dataset Deprecation Framework that includes considerations of risk, mitigation of impact, appeal mechanisms, timeline, post-deprecation protocol, and publication checks that can be adapted and implemented by the machine learning community. Drawing on work on datasheets and checklists, we further offer two sample dataset deprecation sheets and propose a centralized repository that tracks which datasets have been deprecated and could be incorporated into the publication protocols of venues like NeurIPS.

It has been argued that fake news and the spread of false information pose a threat to societies throughout the world, from influencing the results of elections to hindering the efforts to manage the COVID-19 pandemic. To combat this threat, a number of Natural Language Processing (NLP) approaches have been developed. These leverage a number of datasets, feature extraction/selection techniques and machine learning (ML) algorithms to detect fake news before it spreads. While these methods are well-documented, there is less evidence regarding their efficacy in this domain. By systematically reviewing the literature, this paper aims to delineate the approaches for fake news detection that are most performant, identify limitations with existing approaches, and suggest ways these can be mitigated. The analysis of the results indicates that Ensemble Methods using a combination of news content and socially-based features are currently the most effective. Finally, it is proposed that future research should focus on developing approaches that address generalisability issues (which, in part, arise from limitations with current datasets), explainability and bias.

Adversarial attacks against commercial black-box speech platforms, including cloud speech APIs and voice control devices, have received little attention until recent years. The current "black-box" attacks all heavily rely on the knowledge of prediction/confidence scores to craft effective adversarial examples, which can be intuitively defended by service providers without returning these messages. In this paper, we propose two novel adversarial attacks in more practical and rigorous scenarios. For commercial cloud speech APIs, we propose Occam, a decision-only black-box adversarial attack, where only final decisions are available to the adversary. In Occam, we formulate the decision-only AE generation as a discontinuous large-scale global optimization problem, and solve it by adaptively decomposing this complicated problem into a set of sub-problems and cooperatively optimizing each one. Our Occam is a one-size-fits-all approach, which achieves 100% success rates of attacks with an average SNR of 14.23dB, on a wide range of popular speech and speaker recognition APIs, including Google, Alibaba, Microsoft, Tencent, iFlytek, and Jingdong, outperforming the state-of-the-art black-box attacks. For commercial voice control devices, we propose NI-Occam, the first non-interactive physical adversarial attack, where the adversary does not need to query the oracle and has no access to its internal information and training data. We combine adversarial attacks with model inversion attacks, and thus generate the physically-effective audio AEs with high transferability without any interaction with target devices. Our experimental results show that NI-Occam can successfully fool Apple Siri, Microsoft Cortana, Google Assistant, iFlytek and Amazon Echo with an average SRoA of 52% and SNR of 9.65dB, shedding light on non-interactive physical attacks against voice control devices.