Deep neural networks often suffer from overconfidence which can be partly remedied by improved out-of-distribution detection. For this purpose, we propose a novel approach that allows for the generation of out-of-distribution datasets based on a given in-distribution dataset. This new dataset can then be used to improve out-of-distribution detection for the given dataset and machine learning task at hand. The samples in this dataset are with respect to the feature space close to the in-distribution dataset and therefore realistic and plausible. Hence, this dataset can also be used to safeguard neural networks, i.e., to validate the generalization performance. Our approach first generates suitable representations of an in-distribution dataset using an autoencoder and then transforms them using our novel proposed Soft Brownian Offset method. After transformation, the decoder part of the autoencoder allows for the generation of these implicit out-of-distribution samples. This newly generated dataset then allows for mixing with other datasets and thus improved training of an out-of-distribution classifier, increasing its performance. Experimentally, we show that our approach is promising for time series using synthetic data. Using our new method, we also show in a quantitative case study that we can improve the out-of-distribution detection for the MNIST dataset. Finally, we provide another case study on the synthetic generation of out-of-distribution trajectories, which can be used to validate trajectory prediction algorithms for automated driving.

We argue that the phenomena of distributed responsibility, induced acceptance, and acceptance through ignorance constitute instances of imperfect delegation when tasks are delegated to computationally-driven systems. Imperfect delegation challenges human accountability. We hold that both direct public accountability via public transparency and indirect public accountability via transparency to auditors in public organizations can be both instrumentally ethically valuable and required as a matter of deontology from the principle of democratic self-government. We analyze the regulatory content of 16 guideline documents about the use of AI in the public sector, by mapping their requirements to those of our philosophical account of accountability, and conclude that while some guidelines refer to processes that amount to auditing, it seems that the debate would benefit from more clarity about the nature of the entitlement of auditors and the goals of auditing, also in order to develop ethically meaningful standards with respect to which different forms of auditing can be evaluated and compared.

Last week, the United States Senate played host to a number of social media company VPs during hearings on the potential dangers presented by algorithmic bias and amplification. While that meeting almost immediately broke down into a partisan circus of grandstanding grievance airing, Democratic senators did manage to focus a bit on how these recommendation algorithms might contribute to the spread of online misinformation and extremist ideologies. The issues and pitfalls presented by social algorithms are well-known and have been well-documented. So, really, what are we going to do about it? "So I think in order to answer that question, there's something critical that needs to happen: we need more independent researchers being able to analyze platforms and their behavior," Dr. Brandie Nonnecke, Director of the CITRIS Policy Lab at UC Berkeley, told Engadget. Social media companies "know that they need to be more transparent in what's happening on their platforms, but I'm of the firm belief that, in order for that transparency to be genuine, there needs to be collaboration between the platforms and independent peer reviewed, empirical research."

Artificial intelligence is used in a wide range of industries today. One particular implementation of AI that has made headlines recently is in relation to space exploration. NASA reports that they have partnered up with Google, IBM, and Intel in order to develop advanced artificial intelligence technologies that would help drive the future of space sciences. In this post, we take a closer look at how artificial intelligence will be used to drive the mission to explore space. NASA has already started with these implementations, and we are sure to see several new technologies arise from this partnership in the near future.

Google is hosting adverts for unofficial services like applying for a visa that charge vastly inflated amounts, a BBC investigation has revealed. The search engine returned ads for services charging £50 to change an address on a driving licence – something that can be done for free on the government's website. Applying for an ESTA travel permit on the US government's website should cost no more than £10 ($14). But Google'repeatedly' allowed ads for websites charging more than £58 ($80) for an ESTA, the BBC found. Adverts for unofficial services selling government documents are against Google's own rules.

We've been hearing how "data is the new oil" for more than a decade now, and in certain sectors, it's a maxim that has more than panned out. From marketing and logistics to finance and product, decision-making is now dominated by data at all levels of most big private orgs (and if it isn't, I'd be getting a résumé put together, stat). So it might be a something of a surprise to learn that data, which could transform how we respond to the increasingly deadly disasters that regularly plague us, has been all but absent from much of emergency response this past decade. Far from being a geyser of digital oil, disaster response agencies and private organizations alike have for years tried to swell the scope and scale of the data being inputted into disaster response, with relatively meager results. That's starting to change though, mostly thanks to the internet of things (IoT), and frontline crisis managers today increasingly have the data they need to make better decisions across the resilience, response, and recovery cycle.

In 2019, the border town of Chula Vista, about 15 minutes from Tijuana, became California's first " Welcoming City," highlighting the city's financial and educational opportunities for immigrants. It's also one of the nation's most surveilled cities, where the police department uses license plate readers, drones, and body cameras to track residents and has explored facial-recognition technology. Now, those distinctions are clashing, as residents and activists accuse city leaders of "betraying" immigrant residents by permitting federal immigration authorities to access data from license plate readers. That's sparked a citywide movement questioning the city's police department, its surveillance apparatus, and its relationship with residents and immigration enforcement. Since 2015, the Chula Vista Police Department has quietly amassed surveillance tools as part of a smart city approach to policing.

A new report from AI research firm Adversa looked at a number of measurements of the adoption of AI systems, from the number and types of research papers on the topic, to government initiatives that aim to provide policy frameworks for the technology. They found that AI is being rapidly adopted but often without the necessary defenses needed to protect AI systems from targeted attacks. So-called adversarial AI attacks include bypassing AI systems, manipulating results, and exfiltrating the data that the model is based on. These sorts of attacks are not yet numerous, but have happened, and will happen with greater frequency, says Eugene Neelou, co-founder and chief technology officer of Adversa. "Although our research corpus is mostly collected from academia, they have attack cases against AI systems such as smart devices, online services, or tech giant's APIs," he says.

AI has become a part of all our lives. Our cars have automatic braking, platforms like Netflix and Spotify have recommendations, and Alexa and Google can search things for us on command, all powered by artificial intelligence. Although this technology comes with a lot of convenience and advantages, people are also concerned about its dangers. Inadequate security and ethical problems are a few examples of the cons that come with AI. In response to these dangers, the European Union has decided to work on a legal framework to regulate the way AI is used.

A total of nine Artificial Intelligence (AI) research and development (R&D) projects by the DOST-Advanced Science and Technology Institute (DOST-ASTI), University of the Philippines Mindanao (UPMin), De La Salle University (DLSU), University of the Philippines Los Baños (UPLB), and Caraga State University (CarSU) were launched by the Philippines' Department of Science and Technology (DOST Philippines) in April 2021. The AI R&D projects ranging from applications in agriculture to the education sector were launched on April 8 by the Department of Science and Technology – Philippine Council for Industry, Energy and Emerging Technology Research and Development (DOST-PCIEERD) to spur growth in the AI industry in the Philippines. "AI is one of our priority areas as it truly can boost the country and usher us to the fourth industrial revolution. As a powerful agent for good, AI can disrupt traditional processes and provide solutions and opportunities that Filipinos can maximize," said DOST-PCIEERD Executive Director Dr. Enrico C. Paringit during the virtual launch. The Autonomous Societally Inspired Mission Oriented Vehicles (ASIMOV) Program, composed of two-component projects, will be implemented by DOST-ASTI and UPMin.