Goto

Collaborating Authors

Results


The United Nations needs to start regulating the 'Wild West' of artificial intelligence

#artificialintelligence

The European Commission recently published a proposal for a regulation on artificial intelligence (AI). This is the first document of its kind to attempt to tame the multi-tentacled beast that is artificial intelligence. "The sun is starting to set on the Wild West days of artificial intelligence," writes Jeremy Kahn. He may have a point. When this regulation comes into effect, it will change the way that we conduct AI research and development.


Is REAL ID A Real Security Solution? 3 Ways It's Designed To Protect You

#artificialintelligence

Soon, your driver's license may not be enough to get you through airport security in the United States. Oct. 1, 2020 is the deadline for U.S. citizens to have REAL ID-compliant state driver's licenses, a requirement passed by Congress in 2005 in the wake of the Sept. 11, 2001, terrorist attacks. Without a compliant driver's license, those who are 18 and over won't be able to board a domestic flight, unless possessing other specific forms of acceptable identification. The thought behind this was that with standardization, it will become a lot harder to forge documents and gain access to aircraft. While the main idea of REAL ID is to better protect U.S. citizens and their identity, there is controversy over the law.


A Border Town Confronts the Reality of Police Surveillance

WIRED

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.


Scientific excellence and diversity at Annual Meeting

Science

When members of the scientific community gathered at the AAAS Annual Meeting in February, they did so in front of laptops and tablets from their home offices and dining tables. They presented over Zoom, submitted questions via chat, and caught up with colleagues over social media. The 2021 AAAS Annual Meeting was unlike any other in the meeting's 187-year history, but the fully virtual setting did not dampen enthusiasm for sharing science in keeping with the “Understanding Diverse Ecosystems” meeting theme. Dozens of scientific sessions shared new research in areas ranging from microbiomes to space travel. More than 40 workshops offered attendees the opportunity to discuss strategies for working in the ecosystems of academia and science policy. Plenary and topical lecturers covered timely topics, including Ruha Benjamin on how technology can deepen inequities, Anthony Fauci on the next steps for COVID-19 response, Mary Gray on research ethics, and Yalidy Matos on immigration policies. “The quality of the speakers was absolutely undeniable, and the diversity of the speakers—across gender, race, region—was just extraordinary,” said Sudip Parikh, chief executive officer of AAAS and executive publisher of the Science family of journals. “That is what our vision of the world looks like in a place where science is done with creativity and innovation and excellence.” Selecting a diverse meeting program is grounded in AAAS's values, but it is not without concerted effort, according to Claire Fraser. Fraser, who served as AAAS president through February and now serves as chair of the AAAS Board of Directors, selected the meeting theme and led the AAAS Meeting Scientific Program Committee, which oversees selection of the meeting's speakers. “The diversity doesn't happen by accident. I think it reflects the very strong commitment on the part of the Scientific Program Committee to make sure that not only is the science presented timely and excellent, but the diversity of speakers and participants is as broad as it possibly can be,” said Fraser, director of the Institute for Genome Sciences at the University of Maryland School of Medicine. Diversity isn't an afterthought—it's a deliberate part of the very first review of potential scientific sessions, according to Andrew Black, chief of staff and chief public affairs officer. When hundreds of volunteer reviewers evaluate the quality of the submissions before sending the best for consideration by the Scientific Program Committee, they are also looking for diversity across many dimensions, Black said. Among those dimensions are diversity of scientific discipline—befitting AAAS's multidisciplinary membership—but also gender, race and ethnicity, geographic diversity, career stage, and type of institution, including all types and sizes of universities, industry, and government. “Who do you see, who do you hear, and what kind of voices are in dialogue with each other? That's part of our assessment process,” said Agustín Fuentes, professor of anthropology at Princeton University and a member of the Scientific Program Committee. The review process offers opportunities for applicants to diversify their sessions. Applicants are often encouraged to look beyond their own networks to add a range of voices to their presentation to best communicate their ideas to the broader scientific community, Fuentes said. “We need to think very carefully in this moment in time about how do we not only redress past biases and discriminatory practices but how do we create a space, a voice, and a suite of presenters that is very inviting to a diverse audience,” Fuentes said. Added Fraser, “What you end up with is even better because you have such broad perspectives represented.” The committee also emphasized the importance of ensuring that a diverse group of decision-makers have a seat at the table. Members of the Scientific Program Committee, who are nominated from across AAAS and its 26 disciplinary sections and approved by the AAAS Board, represent a broad range of groups and perspectives, Fraser said. “What I firmly believe is that you can't come up with a diverse program like we had this year and like we've had in previous years without that diversity in the program committee,” Fraser said. Commitment to diversity across many axes is part of AAAS Annual Meeting history. In the 1950s, AAAS refused to hold meetings in the segregated South. In 1976, under one of AAAS's first female presidents, Margaret Mead, the Annual Meeting was fully accessible to people with disabilities for the first time. According to the AAAS Project on Science, Technology, and Disability, wheelchair ramps were added to the conference hall, programs were made accessible for hearing-impaired and visually impaired attendees, and Mead's presidential address was simultaneously interpreted in sign language. In 1978, AAAS's Board of Directors voted to move the following year's Annual Meeting out of Chicago because Illinois had not ratified the Equal Rights Amendment. In 1993, AAAS moved its 1999 meeting from Denver after Colorado voters adopted a constitutional amendment to deny residents protection from discrimination based on sexual orientation. Leaders at AAAS note that there is always more work to be done in the present and future—both at the Annual Meeting and year-round. AAAS continues to focus on its own systemic transformation in areas of diversity, equity, and inclusion and on the breadth of initiatives in its new Inclusive STEM Ecosystems for Equity & Diversity program, all to ensure that the scientific enterprise reflects the full range of talent. That goal resonated with many 2021 AAAS Annual Meeting speakers, too. A more diverse group of scientists creating artificial intelligence systems can improve those systems, said Ayanna Howard, a roboticist who leads The Ohio State University's College of Engineering, during her topical lecture, “Demystifying AI Through the Lens of Fairness and Bias.” Said Howard, “We as people are diverse and we're different and it makes us unique and beautiful, and our AI systems should be designed in such a way.” Nalini Nadkarni, a University of Utah biologist who delivered a topical lecture on “Forests, the Earth, and Ourselves: Understanding Dynamic Systems Through an Interdisciplinary Lens,” shared how she reaches young girls to let them know that science—and her own scientific specialty—is a space where they can thrive. She and her students created and distributed “Treetop Barbie,” dressing a doll in fieldwork clothes and creating a doll-sized booklet about canopy plants. The Annual Meeting offers a chance to show that science is best when it is for everyone, regardless of background or perspective, whether they're a kid or just a kid at heart. Said Parikh, “The AAAS Annual Meeting is where the pages of Science literally come alive. It's a place where scientists, no matter what discipline or industry they decided to pursue, can pull back and just fall in love with the idea of science again—like we did when we were kids.”


3 No-Brainer Stocks to Buy in Artificial Intelligence

#artificialintelligence

In fiction, artificial intelligence is often associated with intelligent androids or dystopian futures. But in reality, the AI market mainly revolves around crunching large amounts of data to make quick decisions. Demand for these services -- which power analytics tools, driverless cars, voice assistants, and more -- is climbing. The global AI market was already worth $39.9 billion in 2019, according to Grand View Research, but could still grow at a compound annual growth rate of 42.2% between 2020 and 2027. That's why many companies are jumping aboard the AI bandwagon.


Modern Dimension Reduction

arXiv.org Machine Learning

Data are not only ubiquitous in society, but are increasingly complex both in size and dimensionality. Dimension reduction offers researchers and scholars the ability to make such complex, high dimensional data spaces simpler and more manageable. This Element offers readers a suite of modern unsupervised dimension reduction techniques along with hundreds of lines of R code, to efficiently represent the original high dimensional data space in a simplified, lower dimensional subspace. Launching from the earliest dimension reduction technique principal components analysis and using real social science data, I introduce and walk readers through application of the following techniques: locally linear embedding, t-distributed stochastic neighbor embedding (t-SNE), uniform manifold approximation and projection, self-organizing maps, and deep autoencoders. The result is a well-stocked toolbox of unsupervised algorithms for tackling the complexities of high dimensional data so common in modern society. All code is publicly accessible on Github.


Clearview AI uses your online photos to instantly ID you. That's a problem, lawsuit says

#artificialintelligence

Clearview AI has amassed a database of more than 3 billion photos of individuals by scraping sites such as Facebook, Twitter, Google and Venmo. It's bigger than any other known facial-recognition database in the U.S., including the FBI's. The New York company uses algorithms to map the pictures it stockpiles, determining, for example, the distance between an individual's eyes to construct a "faceprint." This technology appeals to law enforcement agencies across the country, which can use it in real time to help determine people's identities. It also has caught the attention of civil liberties advocates and activists, who allege in a lawsuit filed Tuesday that the company's automatic scraping of their images and its extraction of their unique biometric information violate privacy and chill protected political speech and activity.


'They track every move': how US parole apps created digital prisoners

The Guardian

In 2018, William Frederick Keck III pleaded guilty in a court in Manassas, Virginia, to possession with intent to distribute cannabis. He served three months in prison, then began a three-year probation. He was required to wear a GPS ankle monitor before his trial and then to report for random drug tests after his release. Eventually, the state reduced his level of monitoring to scheduled meetings with his parole officer. Finally, after continued good behaviour, Keck's parole officer moved him to Virginia's lowest level of monitoring: an app on his smartphone.


Battling the Weaponizing of AI

#artificialintelligence

"I don't use Facebook anymore," she said. I was leading a usability session for the design of a new mobile app when she stunned me with that statement. It was a few years back, when I was a design research lead at IDEO and we were working on a service design project for a telecommunications company. The design concept we were showing her had a simultaneously innocuous and yet ubiquitous feature -- the ability to log in using Facebook. But the young woman, older than 20, less than 40, balked at that feature and went on to tell me why she didn't trust the social network any more. This session was, of course, in the aftermath of the 2016 Presidential election. An election in which a man who many regarded as a television spectacle at best and grandiose charlatan at worst had just been elected to our highest office. Though now in 2020, our democracy remains intact.


Us vs. Them: A Dataset of Populist Attitudes, News Bias and Emotions

arXiv.org Artificial Intelligence

Computational modelling of political discourse tasks has become an increasingly important area of research in natural language processing. Populist rhetoric has risen across the political sphere in recent years; however, computational approaches to it have been scarce due to its complex nature. In this paper, we present the new $\textit{Us vs. Them}$ dataset, consisting of 6861 Reddit comments annotated for populist attitudes and the first large-scale computational models of this phenomenon. We investigate the relationship between populist mindsets and social groups, as well as a range of emotions typically associated with these. We set a baseline for two tasks related to populist attitudes and present a set of multi-task learning models that leverage and demonstrate the importance of emotion and group identification as auxiliary tasks.