Many works show that node-level predictions of Graph Neural Networks (GNNs) are unrobust to small, often termed adversarial, changes to the graph structure. However, because manual inspection of a graph is difficult, it is unclear if the studied perturbations always preserve a core assumption of adversarial examples: that of unchanged semantic content. To address this problem, we introduce a more principled notion of an adversarial graph, which is aware of semantic content change. Using Contextual Stochastic Block Models (CSBMs) and real-world graphs, our results uncover: $i)$ for a majority of nodes the prevalent perturbation models include a large fraction of perturbed graphs violating the unchanged semantics assumption; $ii)$ surprisingly, all assessed GNNs show over-robustness - that is robustness beyond the point of semantic change. We find this to be a complementary phenomenon to adversarial examples and show that including the label-structure of the training graph into the inference process of GNNs significantly reduces over-robustness, while having a positive effect on test accuracy and adversarial robustness. Theoretically, leveraging our new semantics-aware notion of robustness, we prove that there is no robustness-accuracy tradeoff for inductively classifying a newly added node.

An antibiogram is a periodic summary of antibiotic resistance results of organisms from infected patients to selected antimicrobial drugs. Antibiograms help clinicians to understand regional resistance rates and select appropriate antibiotics in prescriptions. In practice, significant combinations of antibiotic resistance may appear in different antibiograms, forming antibiogram patterns. Such patterns may imply the prevalence of some infectious diseases in certain regions. Thus it is of crucial importance to monitor antibiotic resistance trends and track the spread of multi-drug resistant organisms. In this paper, we propose a novel problem of antibiogram pattern prediction that aims to predict which patterns will appear in the future. Despite its importance, tackling this problem encounters a series of challenges and has not yet been explored in the literature. First of all, antibiogram patterns are not i.i.d as they may have strong relations with each other due to genomic similarities of the underlying organisms. Second, antibiogram patterns are often temporally dependent on the ones that are previously detected. Furthermore, the spread of antibiotic resistance can be significantly influenced by nearby or similar regions. To address the above challenges, we propose a novel Spatial-Temporal Antibiogram Pattern Prediction framework, STAPP, that can effectively leverage the pattern correlations and exploit the temporal and spatial information. We conduct extensive experiments on a real-world dataset with antibiogram reports of patients from 1999 to 2012 for 203 cities in the United States. The experimental results show the superiority of STAPP against several competitive baselines.

Poverty maps derived from satellite imagery are increasingly used to inform high-stakes policy decisions, such as the allocation of humanitarian aid and the distribution of government resources. Such poverty maps are typically constructed by training machine learning algorithms on a relatively modest amount of ``ground truth" data from surveys, and then predicting poverty levels in areas where imagery exists but surveys do not. Using survey and satellite data from ten countries, this paper investigates disparities in representation, systematic biases in prediction errors, and fairness concerns in satellite-based poverty mapping across urban and rural lines, and shows how these phenomena affect the validity of policies based on predicted maps. Our findings highlight the importance of careful error and bias analysis before using satellite-based poverty maps in real-world policy decisions.

Fears about AI-mediated communication are grounded in different expectations for one's own versus others' use Contribution statement: MJ and NK conceived of the original idea and carried out Study 1. MJ created the stimuli for Studies 1 and 2 and performed the initial analysis of the Study 1 results. ZP was the principal investigator and led the design of Study 2 and its conceptual integration with Study 1 in consultation with all authors. ZP carried out Study 2 in consultation with all authors. MD oversaw ethical approval and pre-registration for Study 2. ZP ran the power analysis, conducted the analyses, designed the figures, and drafted the manuscript in consultation with MD and AM. All authors provided critical feedback and helped shape the design, analysis, and manuscript. Abstract The rapid development of AI-mediated communication technologies (AICTs) - digital tools that use AI to augment our interpersonal messages - has created concerns about the future of interpersonal trust and prompted urgent discussions about disclosure and uptake. We contribute to this discussion by assessing perceptions about acceptability and use of open and secret AICTs for oneself and others. In two studies with representative samples (UK: N=477, US: N=765), we found (a) that secret (i.e., undisclosed) AICT use is deemed less acceptable than open AICT use, (b) people overestimate others' AICT use, and (c) people expect others to use AICTs irresponsibly. Thus, we raise concerns about the potential for misperceptions and different expectations for others to drive self-fulfilling pessimistic outlooks about AImediated communication.

Fox News medical contributor Dr. Marc Siegel weighs in on a survey that found a majority of patients prefer interacting with chatbots over doctors on'Fox News Live.' While 76% of voters want the federal government to regulate artificial intelligence technology, a new Fox News survey finds that only 39% think Uncle Sam is up to the job. In a broader sense, more voters think AI is generally a bad thing, but the view is more positive among those who are familiar with the technology. By an 8-point margin, voters overall are more likely to believe AI is a bad thing for society than a good thing. However, those familiar with AI are more likely to say it's a good thing by a 6-point margin.

Since its establishment in 1914, the US Federal Trade Commission has stood as a bulwark against the fraud, deception, and shady dealings that American consumers face every day -- fining brands that "review hijack" Amazon listings, making it easier to cancel magazine subscriptions and blocking exploitative ad targeting. On Monday, Michael Atleson, Attorney, FTC Division of Advertising Practices, laid out both the commission's reasoning for how emerging generative AI systems like ChatGPT, Dall-E 2 could be used to violate the FTC Act's spirit of unfairness, and what it would do to companies found in violation. "Under the FTC Act, a practice is unfair if it causes more harm than good," Atleson said. "It's unfair if it causes or is likely to cause substantial injury to consumers that is not reasonably avoidable by consumers and not outweighed by countervailing benefits to consumers or to competition." He notes that the new generation of chatbots like Bing, Bard and ChatGPT can be used to influence the user's, "beliefs, emotions, and behavior."

The Biden administration is preparing to examine how companies use artificial intelligence to monitor and manage workers. According to Bloomberg, the White House will publish a blog post later today that invites American workers to share how automated tools are being used in their workplaces. "While these technologies can benefit both workers and employers in some cases, they can also create serious risks to workers," the post states, per Bloomberg. "The constant tracking of performance can push workers to move too fast on the job, posing risks to their safety and mental health." Citing media reports, the White House adds the technology has also been used to deter workers from organizing their workplaces and to perpetuate pay and discipline discrimination.

This Labour Day, working people around the world have little to celebrate. Amid climate change, war and pandemics, inequality is rising, wages are stagnating or even falling, and inflation is skyrocketing, leaving billions of people struggling to make ends meet. In France, where in 1889 labour unions and socialists first designated May 1 as International Workers' Day, hundreds of thousands are demonstrating against a pension law raising the retirement age to 64. However, the sad fact is that many of these workers out on French streets and throughout the world today may not have a job at all come retirement. The speed with which automation and artificial intelligence (AI) are replacing humans in the workplace is breathtaking and poses an unprecedented risk of major economic disruptions and social upheavals.

MASSIVE MANHUNT - Illegal immigrant on the run after allegedly killing five in Texas had been deported five times. Continue reading … BID ACCEPTED - FDIC approves purchase of First Republic Bank by JPMorgan Chase. MORE TO UNCOVER - Antony Blinken'lied boldface to Congress' over contact with Hunter Biden, senator says. 'THERE ARE ONLY TWO GENDERS' - 7th grader speaks out after school allegedly sent him home for controversial t-shirt. EARLY WARNING - Stephen Hawking warned AI could mean the'end of the human race' in years leading up to his death.

Paedophiles are using a popular new artificial intelligence (AI) platform to transform real photos of children into sexualised images, it has been revealed. It has led to warnings to parents to be careful about the pictures of their children they're posting online. The images were found on the US AI image generator Midjourney, which much like ChatGPT uses prompts to deliver an output, although these usually consist of pictures rather than words. The platform is used by millions and has churned out such realistic images that people across the world have been fooled by them, including users on Twitter. An image of Pope Francis donning a huge white puffer jacket with a cross hanging from his neck sent social media users into a frenzy earlier this year.