Speaking in Tongues — Teaching Local Languages to Machines

Curated on Posted on by Stefaan Verhulst

Report by DIAL: “…Machines learn to talk to people by digesting digital content in languages people speak through a technique called Natural Language Processing (NLP). As things stand, only about 85 of the world’s approximately 7500 languages are represented in the major NLPs — and just 7 languages, with English being the most advanced, comprise the majority of the world’s digital knowledge corpus. Fortunately, many initiatives are underway to fill this knowledge gap. My new mini-report with Digital Impact Alliance (DIAL) highlights a few of them from Serbia, India, Estonia, and Africa.

The examples in the report are just a subset of initiatives on the ground to make digital services accessible to people in their local languages. They are a cause for excitement and hope (tempered by realistic expectations). A few themes across the initiatives include –

  • Despite the excitement and enthusiasm, most of the programs above are still at a very nascent stage — many may fail, and others will require investment and time to succeed. While countries such as India have initiated formal national NLP programs (one that is too early to assess), others such as Serbia have so far taken a more ad hoc approach.
  • Smaller countries like Estonia recognize the need for state intervention as the local population isn’t large enough to attract private sector investment. Countries will need to balance their local, cultural, and political interests against commercial realities as languages become digital or are digitally excluded.
  • Community engagement is an important component of almost all initiatives. India has set up a formal crowdsourcing program; other programs in Africa are experimenting with elements of participatory design and crowd curation.
  • While critics have accused ChatGPT and others of paying contributors from the global south very poorly for their labeling and other content services; it appears that many initiatives in the south are beginning to dabble with payment models to incentivize crowdsourcing and sustain contributions from the ground.
  • The engagement of local populations can ensure that NLP models learn appropriate cultural nuances, and better embody local social and ethical norms…(More)”.

AI translation is jeopardizing Afghan asylum claims

Curated on Posted on by Stefaan Verhulst

Article by Andrew Deck: “In 2020, Uma Mirkhail got a firsthand demonstration of how damaging a bad translation can be.

A crisis translator specializing in Afghan languages, Mirkhail was working with a Pashto-speaking refugee who had fled Afghanistan. A U.S. court had denied the refugee’s asylum bid because her written application didn’t match the story told in the initial interviews.

In the interviews, the refugee had first maintained that she’d made it through one particular event alone, but the written statement seemed to reference other people with her at the time — a discrepancy large enough for a judge to reject her asylum claim.

After Mirkhail went over the documents, she saw what had gone wrong: An automated translation tool had swapped the “I” pronouns in the woman’s statement to “we.”

Mirkhail works with Respond Crisis Translation, a coalition of over 2,500 translators that provides interpretation and translation services for migrants and asylum seekers around the world. She told Rest of World this kind of small mistake can be life-changing for a refugee. In the wake of the Taliban’s return to power in Afghanistan, there is an urgent demand for crisis translators working in languages such as Pashto and Dari. Working alongside refugees, these translators can help clients navigate complex immigration systems, including drafting immigration forms such as asylum applications. But a new generation of machine translation tools is changing the landscape of this field — and adding a new set of risks for refugees…(More)”.

The Coming Age of AI-Powered Propaganda

Curated on Posted on by Stefaan Verhulst

Essay by Josh A. Goldstein and Girish Sastry: “In the seven years since Russian operatives interfered in the 2016 U.S. presidential election, in part by posing as Americans in thousands of fake social media accounts, another technology with the potential to accelerate the spread of propaganda has taken center stage: artificial intelligence, or AI. Much of the concern has focused on the risks of audio and visual “deepfakes,” which use AI to invent images or events that did not actually occur. But another AI capability is just as worrisome. Researchers have warned for years that generative AI systems trained to produce original language—“language models,” for short—could be used by U.S. adversaries to mount influence operations. And now, these models appear to be on the cusp of enabling users to generate a near limitless supply of original text with limited human effort. This could improve the ability of propagandists to persuade unwitting voters, overwhelm online information environments, and personalize phishing emails. The danger is twofold: not only could language models sway beliefs; they could also corrode public trust in the information people rely on to form judgments and make decisions.

The progress of generative AI research has outpaced expectations. Last year, language models were used to generate functional proteins, beat human players in strategy games requiring dialogue, and create online assistants. Conversational language models have come into wide use almost overnight: more than 100 million people used OpenAI’s ChatGPT program in the first two months after it was launched, in December 2022, and millions more have likely used the AI tools that Google and Microsoft introduced soon thereafter. As a result, risks that seemed theoretical only a few years ago now appear increasingly realistic. For example, the AI-powered “chatbot” that powers Microsoft’s Bing search engine has shown itself to be capable of attempting to manipulate users—and even threatening them.

As generative AI tools sweep the world, it is hard to imagine that propagandists will not make use of them to lie and mislead…(More)”.

Harnessing Data Innovation for Migration Policy: A Handbook for Practitioners

Curated on Posted on by Stefaan Verhulst

Report by IOM: “The Practitioners’ Handbook provides first-hand insights into why and how non-traditional data sources can contribute to better understanding migration-related phenomena. The Handbook aims to (a) bridge the practical and technical aspects of using data innovations in migration statistics, (a) demonstrate the added value of using new data sources and innovative methodologies to analyse key migration topics that may be hard to fully grasp using traditional data sources, and (c) identify good practices in addressing issues of data access and collaboration with multiple stakeholders (including the private sector), ethical standards, and security and data protection issues…(More)” See also Big Data for Migration Alliance.

What AI Means For Animals

Curated on Posted on by Stefaan Verhulst

Article by Peter Singer and Tse Yip Fai: “The ethics of artificial intelligence has attracted considerable attention, and for good reason. But the ethical implications of AI for billions of nonhuman animals are not often discussed. Given the severe impacts some AI systems have on huge numbers of animals, this lack of attention is deeply troubling.

As more and more AI systems are deployed, they are beginning to directly impact animals in factory farms, zoos, pet care and through drones that target animals. AI also has indirect impacts on animals, both good and bad — it can be used to replace some animal experiments, for example, or to decode animal “languages.” AI can also propagate speciesist biases — try searching “chicken” on any search engine and see if you get more pictures of living chickens or dead ones. While all of these impacts need ethical assessment, the area in which AI has by far the most significant impact on animals is factory farming. The use of AI in factory farms will, in the long run, increase the already huge number of animals who suffer in terrible conditions.

AI systems in factory farms can monitor animals’ body temperature, weight and growth rates and detect parasites, ulcers and injuries. Machine learning models can be created to see how physical parameters relate to rates of growth, disease, mortality and — the ultimate criterion — profitability. The systems can then prescribe treatments for diseases or vary the quantity of food provided. In some cases, they can use their connected physical components to act directly on the animals, emitting sounds to interact with them — giving them electric shocks (when the grazing animal reaches the boundary of the desired area, for example), marking and tagging their bodies or catching and separating them.

You might be thinking that this would benefit the animals — that it means they will get sick less often, and when they do get sick, the problems will be quickly identified and cured, with less room for human error. But the short-term animal welfare benefits brought about by AI are, in our view, clearly outweighed by other consequences…(More)” See also: AI Ethics: The Case for Including Animals.

Including the underrepresented

Curated on Posted on by Stefaan Verhulst

Paper by FIDE: “Deliberative democracy is based on the premise that all voices matter and that we can equally participate in decision-making. However, structural inequalities might prevent certain groups from being recruited for deliberation, skewing the process towards the socially privileged. Those structural inequalities are also present in the deliberation room, which can lead to unconscious (or conscious) biases that hinder certain voices while amplifying others. This causes particular perspectives to influence decision-making unequally.

This paper presents different methods and strategies applied in previous processes to increase the inclusion of underrepresented groups. We distinguish strategies for the two critical phases of the deliberative process: recruitment and deliberation…(More)”.

The Surveillance Ad Model Is Toxic — Let’s Not Install Something Worse

Curated on Posted on by Stefaan Verhulst

Article by Elizabeth M. Renieris: “At this stage, law and policy makers, civil society and academic researchers largely agree that the existing business model of the Web — algorithmically targeted behavioural advertising based on personal data, sometimes also referred to as surveillance advertising — is toxic. They blame it for everything from the erosion of individual privacy to the breakdown of democracy. Efforts to address this toxicity have largely focused on a flurry of new laws (and legislative proposals) requiring enhanced notice to, and consent from, users and limiting the sharing or sale of personal data by third parties and data brokers, as well as the application of existing laws to challenge ad-targeting practices.

In response to the changing regulatory landscape and zeitgeist, industry is also adjusting its practices. For example, Google has introduced its Privacy Sandbox, a project that includes a planned phaseout of third-party cookies from its Chrome browser — a move that, although lagging behind other browsers, is nonetheless significant given Google’s market share. And Apple has arguably dealt one of the biggest blows to the existing paradigm with the introduction of its AppTrackingTransparency (ATT) tool, which requires apps to obtain specific, opt-in consent from iPhone users before collecting and sharing their data for tracking purposes. The ATT effectively prevents apps from collecting a user’s Identifier for Advertisers, or IDFA, which is a unique Apple identifier that allows companies to recognize a user’s device and track its activity across apps and websites.

But the shift away from third-party cookies on the Web and third-party tracking of mobile device identifiers does not equate to the end of tracking or even targeted ads; it just changes who is doing the tracking or targeting and how they go about it. Specifically, it doesn’t provide any privacy protections from first parties, who are more likely to be hegemonic platforms with the most user data. The large walled gardens of Apple, Google and Meta will be less impacted than smaller players with limited first-party data at their disposal…(More)”.

Innovating Democracy? The Means and Ends of Citizen Participation in Latin America

Curated on Posted on by Stefaan Verhulst

Book by Thamy Pogrebinschi: “Since democratization, Latin America has experienced a surge in new forms of citizen participation. Yet there is still little comparative knowledge on these so-called democratic innovations. This Element seeks to fill this gap. Drawing on a new dataset with 3,744 cases from 18 countries between 1990 and 2020, it presents the first large-N cross-country study of democratic innovations to date. It also introduces a typology of twenty kinds of democratic innovations, which are based on four means of participation, namely deliberation, citizen representation, digital engagement, and direct voting. Adopting a pragmatist, problem-driven approach, this Element claims that democratic innovations seek to enhance democracy by addressing public problems through combinations of those four means of participation in pursuit of one or more of five ends of innovations, namely accountability, responsiveness, rule of law, social equality, and political inclusion…(More)”.

Soft power, hard choices: Science diplomacy and the race for solutions

Curated on Posted on by Stefaan Verhulst

Article by Stephan Kuster and Marga Gual Soler: “…Global challenges demand that we build consensus for action. But reaching agreement on how – and even if – science and technology should be applied, for the aggregate benefit of all, is complex, and increasingly so.

Science and technology are tightly intertwined with fast-changing economic, geopolitical, and ideological agendas. That pace of change complicates, and sometimes deviates, the discussions and decisions that could unlock the positive global impact of scientific advances.

Therefore, anticipation is key. Understanding the societal, economic, and geopolitical consequences of emerging and possible new technologies before they are deployed is critical. Just recently, for example, artificial intelligence (AI) labs have been urged by a large number of researchers and leading industry figures to pause the training of powerful AI systems, given the inherent risks to society and humanity’s existence.

Indeed, the rapid pace of scientific development calls for more effective global governance when it comes to emerging technology. That in turn requires better anticipatory tools and new mechanisms to embed the science community as key stakeholder and influencer in this work.

The Geneva Science and Diplomacy Anticipator (GESDA) was created with those goals in mind. GESDA identifies the most significant science breakthroughs in the next five, 10, and 25 years. It assesses those advances with the potential to most profoundly to impact people, society, and the planet. It then brings together scientific and policy leaders from around the world to devise the diplomatic envelopes and approaches needed to embrace these advances, while minimizing downsides risks of unintended consequences…(More)”.

The Technology/Jobs Puzzle: A European Perspective

Curated on Posted on by Stefaan Verhulst

Blog by Pierre-Alexandre Balland, Lucía Bosoer and Andrea Renda as part of the work of the Markle Technology Policy and Research Consortium: “In recent years, the creation of “good jobs” – defined as occupations that provide a middle-class living standard, adequate benefits, sufficient economic security, personal autonomy, and career prospects (Rodrik and Sabel 2019; Rodrik and Stantcheva 2021) – has become imperative for many governments. At the same time, developments in industrial value chains and in digital technologies such as Artificial Intelligence (AI) create important challenges for the creation of good jobs. On the one hand, future good jobs may not be found only in manufacturing, ad this requires that industrial policy increasingly looks at services. On the other hand, AI has shown the potential to automate both routine and also non-routine tasks (TTC 2022), and this poses new, important questions on what role humans will play in the industrial value chains of the future. In the report drafted for the Markle Technology Policy and Research Consortium on The Technology/Jobs Puzzle: A European Perspective, we analyze Europe’s approach to the creation of “good jobs”. By mapping Europe’s technological specialization, we estimate in which sectors good jobs are most likely to emerge, and assess the main opportunities and challenges Europe faces on the road to a resilient, sustainable and competitive future economy.The report features an important reflection on how to define job quality and, relatedly “good jobs”. From the perspective of the European Union, job quality can be defined along two distinct dimensions. First, while the internationally agreed definition is rather static (e.g. related to the current conditions of the worker), the emerging interpretation at the EU level incorporates the extent to which a given job leads to nurturing human capital, and thereby empowering workers with more skills and well-being over time. Second, job quality can be seen from a “micro” perspective, which only accounts for the condition of the individual worker; or from a more “macro” perspective, which considers whether the sector in which the job emerges is compatible with the EU’s agenda, and in particular with the twin (green and digital) transition. As a result, we argue that ideally, Europe should avoid creating “good” jobs in “bad” sectors, as well as “bad” jobs in “good” sectors. The ultimate goal is to create “good” jobs in “good” sectors….(More)”