Drivers in the UK are taking legal action against Uber over its real-time biometric identification checks, TechCrunch has reported. A union representing the drivers claims that some members were wrongly suspended when they were misidentified, and lost their licenses as a result. "[The] facial recognition systems... are inherently faulty and generate particularly poor accuracy results when used with people of color," the App Drivers & Couriers Union (ADCU) wrote in a blog post. Backed by two worker's rights groups, the union is crowdfunding the legal action, taken on behalf of former UberEats courier Pa Edrissa Manjang and former Uber driver Imran Javaid Raja. It said they were "unfairly dismissed after the company's facial recognition system failed to identify them." "Workers are prompted to provide a real-time selfie and face dismissal if the system fails to match the selfie with a stored reference photo," the ADCU wrote.

In a recently published paper, a pair of academics propose that the application of artificial intelligence can offer a potent weapon against antitrust behavior in the Big Tech sector. This is the very industry that has advanced this technology, noted one of those academics, Giovana Massarotto, a Center for Technology, Innovation and Competition academic fellow at the University of Pennsylvania Carey Law School and an adjunct professor at the University of Iowa. She underscored this fact in an article for Bloomberg Law, in which she maintains that "the present economic democracy propaganda against Big Tech is not the solution to increase competition in fast-moving technology markets." In fact, she says, the industry's ingenuity is needed to achieve our nation's pro-competition goals. Massarotto and University of Liege (Belgium) Associate Professor Ashwin Ittoo write about their "antitrust machine learning application" (AML) which shows the potential for AI to "assist antitrust agencies in detecting anticompetitive practices faster."

Synthetic datasets are becoming increasingly popular for training artificial intelligence models. Proponents of this computer-generated data say it protects personal information and reduces the chances of bias emerging in AI systems. But for many, concerns over privacy and accuracy remain. New use cases for synthetic data are emerging daily. On Thursday, the International Organization for Migration (IOM) charity announced the launch of a synthetic dataset for human trafficking, which has been developed in partnership with Microsoft Research.

We study the privacy implications of deploying recurrent neural networks in machine learning. We consider membership inference attacks (MIAs) in which an attacker aims to infer whether a given data record has been used in the training of a learning agent. Using existing MIAs that target feed-forward neural networks, we empirically demonstrate that the attack accuracy wanes for data records used earlier in the training history. Alternatively, recurrent networks are specifically designed to better remember their past experience; hence, they are likely to be more vulnerable to MIAs than their feed-forward counterparts. We develop a pair of MIA layouts for two primary applications of recurrent networks, namely, deep reinforcement learning and sequence-to-sequence tasks. We use the first attack to provide empirical evidence that recurrent networks are indeed more vulnerable to MIAs than feed-forward networks with the same performance level. We use the second attack to showcase the differences between the effects of overtraining recurrent and feed-forward networks on the accuracy of their respective MIAs. Finally, we deploy a differential privacy mechanism to resolve the privacy vulnerability that the MIAs exploit. For both attack layouts, the privacy mechanism degrades the attack accuracy from above 80% to 50%, which is equal to guessing the data membership uniformly at random, while trading off less than 10% utility.

The premise of this paper is that compliance with Trustworthy AI governance best practices and regulatory frameworks is an inherently fragmented process spanning across diverse organizational units, external stakeholders, and systems of record, resulting in process uncertainties and in compliance gaps that may expose organizations to reputational and regulatory risks. Moreover, there are complexities associated with meeting the specific dimensions of Trustworthy AI best practices such as data governance, conformance testing, quality assurance of AI model behaviors, transparency, accountability, and confidentiality requirements. These processes involve multiple steps, hand-offs, re-works, and human-in-the-loop oversight. In this paper, we demonstrate that process mining can provide a useful framework for gaining fact-based visibility to AI compliance process execution, surfacing compliance bottlenecks, and providing for an automated approach to analyze, remediate and monitor uncertainty in AI regulatory compliance processes.

Japanese-American scientist Syukuro Manabe, Klaus Hasselmann of Germany and Giorgio Parisi of Italy on Tuesday won the Nobel Physics Prize for climate models and the understanding of physical systems. The Nobel committee said it was sending a message with its prize announcement just weeks before the COP26 climate summit in Glasgow, as the rate of global warming sets off alarm bells around the world. "The world leaders that haven't got the message yet, I'm not sure they will get it because we are saying it," said Thor Hans Hansson, chair of the Nobel Committee for Physics. "But … what we are saying is that the modeling of climate is solidly based in physics theory." Manabe, 90, and Hasselmann, 89, will share half of the 10 million kronor ($1.1 million) prize for their research on climate models.

An Uber driver who lost his job when automated face-scanning software failed to recognise him is accusing the firm of indirect race discrimination in a legal test case. The black driver, who worked on the Uber platform from 2016 until April 2021, has filed an employment tribunal claim alleging his account was illegally deactivated when facial-verification software used to log drivers on to the ride-hailing app decided he wasn't who he said he was. The Independent Workers of Great Britain trade union, which is backing the action, claimed at least 35 other drivers had had their registration with Uber terminated as a result of alleged mistakes with the software since the start of the pandemic. It is calling for Uber to scrap the "racist algorithm" and reinstate terminated drivers. An Uber spokeswoman said the firm "strongly refutes the completely unfounded claims" and said it is "committed to fighting racism and being a champion for equality--both inside and outside our company."

Fox News Flash top headlines are here. Check out what's clicking on Foxnews.com. Non-White employees for the City of Seattle were invited to an anti-racist training to explore their internalized racism and given tools on how to free themselves from White supremacy influence, new documents show. Employees "who identify as people of color" were sent an invite to a three-hour optional training, "Internalized Racial Inferiority," in August that excluded White people. It was conducted on Sept. 3 by the city's Race and Social Justice Initiative, radio host Jason Rantz on KTTH reported Monday after filing a public disclosure request for the lesson materials.

We characterize optimal oversight of algorithms in a world where an agent designs a complex prediction function but a principal is limited in the amount of information she can learn about the prediction function. We show that limiting agents to prediction functions that are simple enough to be fully transparent is inefficient as long as the bias induced by misalignment between principal's and agent's preferences is small relative to the uncertainty about the true state of the world. Algorithmic audits can improve welfare, but the gains depend on the design of the audit tools. Tools that focus on minimizing overall information loss, the focus of many post-hoc explainer tools, will generally be inefficient since they focus on explaining the average behavior of the prediction function rather than sources of mis-prediction, which matter for welfare-relevant outcomes. Targeted tools that focus on the source of incentive misalignment, e.g., excess false positives or racial disparities, can provide first-best solutions. We provide empirical support for our theoretical findings using an application in consumer lending.

In a nutshell, any managers' value can be correlated to their team's efficiency and productivity. It can be argued that besides using sophisticated strategies, many managers also rely on their gut feeling to assess employees and treat them accordingly. With the pandemic forcing remote work upon the majority of businesses, this approach has become largely irrelevant. It has become increasingly hard for managers to monitor employee productivity without direct communication, which dramatically increased the demand for remote surveillance solutions. However, the notion of advanced employee monitoring has been around for ages.