Vets

'Ethics of using artificial intelligence (AI) in veterinary medicine' by Simon Coghlan and Thomas Quinn in (2023) AI and Society comments 

This paper provides the first comprehensive analysis of ethical issues raised by artificial intelligence (AI) in veterinary medicine for companion animals. Veterinary medicine is a socially valued service, which, like human medicine, will likely be significantly affected by AI. Veterinary AI raises some unique ethical issues because of the nature of the client–patient–practitioner relationship, society’s relatively minimal valuation and protection of nonhuman animals and differences in opinion about responsibilities to animal patients and human clients. The paper examines how these distinctive features influence the ethics of AI systems that might benefit clients, veterinarians and animal patients—but also harm them. It offers practical ethical guidance that should interest ethicists, veterinarians, clinic owners, veterinary bodies and regulators, clients, technology developers and AI researchers. ... 

AI—i.e. digital systems that perform tasks normally requiring human intelligence (Russell and Norvig 2021)—is poised to transform human medicine (Topol 2019; Wilson et al. 2021) and may prove equally transformative of veterinary medicine (Basran and Appleby 2022; WIRED Brand Lab 2022). Like human medical AI (Astromskė et al. 2021; Dalton-Brown 2020; Keskinbora 2019), veterinary AI raises important ethical issues. Although several papers touch on ethical aspects of veterinary AI (Appleby and Basran 2022; Ezanno et al. 2021; Steagall et al. 2021), including its implications for ‘livestock’(Neethirajan 2021),a more detailed ethical evaluation of companion animal AI is wanting. Our analysis of AI’s ethical implications for companion animal medicine should interest ethicists, veterinarians, clinic owners, veterinary bodies and regulators, clients, technology developers and AI researchers. 

Veterinary practice raises unique ethical issues that stem from the client–patient–practitioner relationship. Companion animals are potentially more exposed to harms from AI than are humans because they lack the same strong social, moral and legal status. For example, the law does not effectively protect animals from wrongful injury or from clients who seek unwarranted or unjustified ‘euthanasia’ (Favre 2016). These conditions are relevant to the ethics of veterinary AI. At the same time, medical AI raises its own distinctive ethical issues—issues like trust, data security and algorithmic transparency—which we also discuss in the veterinary context. 

AI in veterinary medicine might be used for business purposes and hospital logistics like booking appointments. Technology that affects practitioner workflow could have ethical implications, as could other AI, such as language translation apps that enable communication with linguistically diverse clients. However, AI for triage, diagnosis, prognosis and treatment raises the most distinctive, complex and consequential ethical questions. We concentrate on AI for such medical decision-making. 

Currently, AI enjoys massive public and private investment, propelled by stories like algorithms defeating Jeopardy and Go masters (Mitchell 2019). Another indication of AI’s rapid ascent are recent large language models like ChatGPT and text-to-image generators that demonstrate remarkable, though sometimes strange and biased, outputs (see Fig. 1). Yet most people are bewildered by the technical jargon of artificial neural networks, deep learning, computer vision, random forests and natural language processing (Waljee and Higgins 2010). Veterinary practitioners too may not always understand, for instance, the ways in which AI learns from data and autonomously updates its algorithms to draw inferences about previously unencountered data (e.g. from patient radiographs or medical records)—and this may create uncertainty about its use in healthcare. 

This issue of trust in technology is important. To some degree, medical AI remains just as much an art as a science (Quinn et al. 2021b), and AI developers are only now exploring how to apply modern machine learning (ML) methods successfully in medicine. This involves experimenting with how data are collected and pre-processed, how AI models are applied and optimised and how model performance is evaluated. Each step contains many nuances that could affect model operation in clinic settings and unintentionally harm patients and clients. While busy practitioners cannot be expected to understand all these nuances, they will increasingly need at least a basic understanding of the ethical risks and benefits of AI. This paper identifies and examines these ethical issues. 

The paper runs as follows. Section 2 outlines medical AI in veterinary practice. Section 3 introduces ethical principles of AI, human medicine and veterinary medicine. Section 4 identifies and examines nine ethical issues raised by veterinary AI. Section 5 discusses important ethical norms in veterinary medicine and AI’s distinctive implications in that realm, as well as providing some practical guidance for AI’s use.

Infrastructure

'Infrastructuring the Digital Public Sphere' by Julie E Cohen in (2023) 25 Yale Journal of Law & Technology (forthcoming) comments 

The idea of a "public sphere"-- a shared, ideologically neutral domain where ideas and arguments may be shared, encountered, and contested -- serves as a powerful imaginary in legal and policy discourse, informing both assumptions about how public communication works and ideals to which inevitably imperfect realities are compared. In debates about feasible and legally permissible content governance mechanisms for digital platforms, the public sphere ideal has counseled attention to questions of ownership and control rather than to other, arguably more pressing questions about systemic configuration. This essay interrogates such debates through the lens of infrastructure, with particular reference to the ways that digital tracking and advertising infrastructures perform systemic content governance functions. 

Social infrastructure of a different sort in recommendations by the Productivity Commission in its Advancing Prosperity: 5-year Productivity Inquiry report.

The Commission states

Building an adaptable workforce: education 

Reflecting the role of education in creating a high skilled and highly adaptable workforce, broad ranging reforms are proposed across higher education, vocational education and training (VET), schools and lifelong learning. These reforms emphasise stronger foundational learning to support further skills acquisition throughout individuals’ working lives via a broader array of flexible options. 

Higher education reforms aim to create a more dynamic university sector, putting greater emphasis on quality teaching. Loan reforms would expand access to high quality VET, and encourage emerging vocational options that develop broad, adaptive and less occupation specific skills. 

A more coherent approach to lifelong learning and ongoing skill development is based on targeted tax incentives, and the improved availability and recognition of flexible, short form training options.  

Long term improvements in school outcomes are possible through increasing (and judicious) use of learning technology and a stronger link between pedagogical evidence and classroom practice. 

Proposed reforms focus on assisting governments and schools in this journey. 

 Reform directive 1: Improve schools’ capacity to lay the educational foundations for the future workforce 

Recommendation 8.1 Leverage digital technology in schools 

State and Territory Governments should work with schools to extend, improve and embed the use of education technology in order to realise future benefits for students. Initiatives should aim to: • enable teaching practices to evolve with the changing classroom environment by prioritising the development and implementation of digital tools to support teaching and learning, while balancing flexibility for individual jurisdictions’ needs – this could include developing an online assessment tool and giving the Australian Education Research Organisation (AERO) responsibility for researching and vetting effective digital technologies to be implemented in schools • replace manual school administrative processes with technology based and automated solutions where this has not been done already – this could include evaluating technology based solutions for administrative processes currently in place and developing mechanisms to diffuse these to other schools • support continuous commitment to ongoing professional development modules that support teachers in using data analytics to drive student improvement. 

Recommendation 8.2 Make best practice teaching common practice 

State and Territory Governments should facilitate greater classroom access for the Australian Education Research Organisation (AERO) to support more principal and teacher involvement in education research to ensure that evidence based research provides information that is salient and readily applicable by practitioners. Initiatives should focus on: • enabling greater observation of, and feedback on, classroom teaching practices, by supporting more informal teacher networks, and creating or strengthening the existing roles within the local school system for highly accomplished and lead teachers (HALT) to share their in depth knowledge and skills with their colleagues • increasing curriculum implementation support for teachers, by curating high quality, evidence based and government endorsed curriculum resources (curriculum plans, whole subject sequences, lesson plans and classroom tools), to be made available for teachers and school leaders from a single source.   

Reform directive 2: Enable innovative schooling approaches for improved learning outcomes 

Recommendation 8.3 Enable experimentation with alternative approaches to schooling 

State and Territory Governments should be open to experimenting with new, innovative school models or operational changes where there is an evidence base (including overseas) to suggest outcomes could be improved for Australian students. In the first instance, legislative, regulatory, administrative or policy barriers that would prevent individual schools varying their operating model should be removed. In addition, there should be capacity and appropriate resourcing within the local school system to allow the merits of any trials to be evaluated. Innovations should aim to: • offer different lesson delivery options to lift quality teaching and learning, including for example, offering online classes in the absence of a teacher with the relevant expertise in a topic, or trials of untimed syllabus approaches to promote a continuous learning process • better cater to student needs to encourage school attendance and lift student outcomes, including through variations in school hours and use of technology to personalise students’ learning environment. 

Reform directive 3: Grow access to tertiary education 

Recommendation 8.4 Grow access to higher education over time 

The Australian Government should adopt an improved demand driven model for providing Commonwealth supported places to domestic undergraduate university students, subject to measures outlined in other recommendations that: contain fiscal costs (recommendation 8.5); and ensure all students are adequately supported (recommendations 8.13 and 8.14).   

Recommendation 8.5 Better targeting of investment in higher education 

The Australian Government should introduce a new university funding model to better target investment while facilitating wider access to higher education. • Total university funding per student by field of study (comprising the student contribution and government contribution) should continue to be the cost of delivery for that field (reflecting a median estimate of efficient costs with the methodology to be refined over time as outlined in recommendation 8.6). • The student contribution should be set based on average expected earnings for each field of study, with students with a greater capacity to repay incurring more debt. Student contributions should be higher, on average, to recoup a greater share of the costs of university from those who benefit from attending university, rather than recouping this from the broader tax base. This would also help to fund the return to a demand driven system. • The government contribution should make up the gap between the student contribution and estimated cost of delivery for each field of study. 

Recommendation 8.6 Improve price setting in tertiary education 

The Australian Government should conduct regular costing exercises to estimate the cost of delivering tertiary teaching and research. The methodology underpinning these cost exercises should be periodically reviewed and refined to inform more accurate cost estimates, and should aim to ultimately reflect only efficient costs. These cost estimates should inform funding as well as price and loan caps, to encourage efficient delivery of quality education and research by tertiary institutions. 

Recommendation 8.7 Expand loan eligibility to more students 

The Australian Government, in consultation with State and Territory governments, should gradually expand VET Student Loan eligibility. • Access should expand to more Diploma and Advanced Diploma level courses. Instead of current criteria, all courses should be eligible except those that are primarily taken for leisure or have demonstrated poor labour market outcomes. This expansion should be evaluated after a suitable period, including observed effects of the earlier expansion on student participation, course decisions and employment outcomes; and any evidence of rorting by providers. Following this evaluation, and addressing any implementation issues, eligibility should also be considered for Certificate IV and Certificate III courses. • Loan fee arrangements should also be equalised across the tertiary sector, levied on all students regardless of type (that is, extended from fee for service VET students and non university higher education students to include subsidised VET students and university students). The loan fee rate should also be lowered reflecting application to a broader base of students. 

Reform directive 4: Support a culture of lifelong learning for an agile workforce 

Recommendation 8.8 Consolidate support for lifelong learning 

The Australian Government should consolidate and examine the effectiveness and accessibility of available programs to support lifelong learning and to reduce gaps and increase uptake. In doing so, it should evaluate the effectiveness of targeted programs to inform and prioritise policies for a consolidated lifelong learning strategy by: • trialling policies that target support at employed lower income people, including vouchers for career planning and work related upskilling and reskilling • evaluating the incoming Skills and Training Boost to assess its effects on the uptake of additional overall training, the skills it develops, productivity, labour mobility, and the characteristics of the businesses most responsive to the measure. Government linked administrative datasets will be useful for such an evaluation but might need to be supplemented • extending the existing capacity for self education deductions to education that is likely to lead to additional income outside of the employee’s existing employment. This change should be evaluated after a suitable period, and pursued subject to assurance that strong integrity measures can effectively reduce the risks of fraudulent claims • examining the effectiveness of training programs delivered to people who are unemployed and those transitioning to work such as Employability Skills Training programs, particularly for people later in life. Government should also increase the accessibility, flexibility, and coherence of available pathways by: • extending income contingent loans to more VET courses (recommendation 8.7) • providing alternative exit opportunities through the provision of nested qualifications (recommendation 8.13) • requiring publicly funded universities to make their lecture materials available online, with consideration of extending this to some aspects of government funded VET where that is practically feasible (recommendation 8.9) • ensuring that the Australian Government’s Microcred Seeker extend beyond courses supplied by TEQSA recognised providers to the VET sector and where possible, to other private and well recognised domestic and international course offerings • constraining regulations that make acquiring new skills and moving to new occupations overly onerous. Most particularly, through regular review of occupational licensing policies and addressing issues in scope of practice (reform directive 10).   

Reform directive 5: Increase tertiary education teaching quality to underpin a well trained workforce 

Recommendation 8.9 Leverage information to improve quality 

The Australian Government should: • increase the transparency of teaching performance by requiring universities to provide all lectures online and for free • refine and validate new Quality Indicators for Learning and Teaching (QILT), and use these and other data to develop and publish more meaningful indicators of tertiary teaching quality and performance • adapt the ComparED tool to address the risk that students may misunderstand its information and consider the option of abandoning it and providing additional QILT data to non government funded websites that cover many other aspects of higher education providers relevant to student choice • give the Tertiary Education Quality and Standards Agency (TEQSA) the responsibility to undertake external university teaching quality assurance review processes akin to those applied by the Quality Assurance Agency (Scotland). 

Recommendation 8.10 Professionalise the teaching role 

The Australian Government should bolster the incentives for, and prestige of, higher education teaching by: • facilitating trials of additional funding for undertaking research and teaching development provided to individual staff based on their teaching performance, drawing on the Griffith Business School’s Teaching Excellence Recognition Scheme (TERS) • trialling a modest Australian Research Council Grant that provides funding for teaching focused research for 6 months to a year • enhancing preparation for higher education teaching, informed by the evidence collected by initiatives outlined in recommendations 8.9 and 8.11. 

Recommendation 8.11 Develop an Australian evidence base 

The Australian Government should extend the role of the Australian Education Research Organisation (AERO) to the collection and dissemination of evidence on best practice post school teaching, covering both VET and higher education. As part of this new role, AERO should also: • draw on the lessons from the teaching practices of awardees of the Australian Government’s Australian Awards for University Teaching • undertake a rapid review of the use of formative and summative review processes and professional development initiatives in higher education institutions. 

Recommendation 8.12 Favour light handed and simple incentives over performance based funding 

The Australian Government should: • put on hold the scheduled commencement of performance based funding of universities in 2024 and only reinstitute if its risks are better managed and if other approaches to improving the performance of universities have proved ineffective • explore the option of financial rewards to higher education providers that AERO identifies as having made successful efforts to improve and use formative assessment tools and professional development (drawing on recommendation 8.11). 

Reform directive 6: Better and more flexible matching between students and work opportunities 

Recommendation 8.13 Expand alternative exit opportunities through the provision of nested qualifications 

The Australian Government should require that for any given undergraduate degree, Australian higher education providers create at least one subset of courses that, if completed, lead to a lower level qualification for students who decide to withdraw before completing the whole degree (‘a nested qualification’). The Australian Government should leave the design, requirements, and timing of the nested qualification/s to providers’ discretion, with the exception that any qualification would need to meet the relevant Tertiary Education Quality and Standards Agency (TEQSA) standards and monitoring requirements. 

Recommendation 8.14 Give students support to complete and clarity to exit 

The Australian Government should amend the Higher Education Support Act 2003 (Cth) (HESA) to support completion where desirable and facilitate early exits where necessary. It should do this by: • providing grants to encourage higher education providers to experiment with and share new strategies for student retention • assessing any individual grant for its effectiveness and lessons in post implementation reviews and evaluating the higher education grant program as a whole after six years to determine whether rounds of funding under the grant have contributed to a demonstrable improvement in student completion rates • amending the ‘census date’ in the HESA to the ‘payment date’ and requiring that universities effectively communicate to students that the payment date is the time when they can exit without having to pay fees for any initially commenced course.

Hart

'The Rule of Law: “A” Relation Between Law and Morals' by Alani Golanski in (2022) 42(2) Northern Illinois University Law Review comments 

H. L. A. Hart begins chapter nine of The Concept of Law by saying that “[t]here are many different types of relation between law and morals and there is nothing which can be profitably singled out for study as the relation between them.”1 He allows that conventional and social group morality, as well as a more rigorously practiced and “enlightened” moral criticism, have both “profoundly influenced” the development of law. Most saliently, legal systems, but also institutions generally, will as a “natural necessity” incorporate a “minimum content of [n]atural [l]aw” in service of the human propensity toward survival. 

This minimum content of natural law includes certain substantive prohibitions. Human beings are vulnerable creatures, of bounded capabilities as well as limited altruism. All are tempted, at least sometimes, to pursue their own immediate interests at the expense of others’ welfare. This is “one of the natural facts which makes the step from merely moral to organized, legal forms of control a necessary one.” So law, like morality, will ordinarily proscribe such offenses as murder or unprovoked assault aimed at appropriating one’s neighbor’s assets. Given the “standing danger” that there are always some who will try to exploit and overcome merely moral constraints, “what reason demands is voluntary cooperation in a coercive system.” 

These sorts of considerations that draw moral values into a relation with law do not warrant the different conclusion that conformity with morality provides a necessary criterion of the existence or validity of law. Rejection of that conclusion is one of Hart’s main legal positivist premises, his “separation thesis.” Nor can there be a necessary correlation, for Hart, between legal rules and natural-law theory’s ample moral standards, because “the purposes men have for living in society are too conflicting and varying to make possible much extension of the argument” that the legal system’s content necessarily encompasses more than natural law’s minimum content. 

Yet, for Hart, the factors that warrant an acknowledgment that legal systems generally must incorporate a minimum content of natural law spill over into his understanding of what he terms “legality.” This concept implicates the manner in which laws come into existence, as well as characteristics of the laws requisite to any legal system’s abilities to effectuate social control. The central claim is that, if the legal system is to fulfill its social control function, and by close analogy abide by legality, the system’s outputs will have to abide by certain formal requirements that bring enacted rules “within the capacity of most to obey.” Hence, the legal rules will have to satisfy certain conditions, such as being intelligible and typically not retrospective. 

The notion of legality is intimately related to, and often taken to be synonymous with, the idea of the rule of law. Hart’s project conduced him toward minimizing the perception of any necessary connection between law and morals, and his minimal expression of legality has influenced slim and formal conceptions of the rule of law ideal itself. This is how Jeremy Waldron put it:

I think Hart was inclined to see a preoccupation with legality and the rule of law as a source of confusion in jurisprudence; often one gets the impression that Hart thought that if anyone offered to talk about it, the responsible thing to do was to say something palliative and then shut down the discussion as quickly and firmly as possible. Principles of legality, Hart implied, may be among the principles we should use for the evaluation of law, but their study is not part of the philosophical discipline that tries to tell us what law essentially is.

That is fair enough. But the impulse criticized by Waldron to sever the focus on the concept of law from an evaluation of law and legal systems should not precondition the criteria by which legal institutional action is evaluated. Limiting the rule of law evaluation to an appraisal of whether the system abides by a few, sharply delimited formal conditions is likely in inevitable tension with the general impulse to evaluate law’s workings more deeply. This does not mean that just any sort of evaluation of the legal system counts as a rule of law assessment. Nor, however, does a robust concept of the rule of law as an exercise in political morality impair a “hard” legal positivist concept of law that emphatically excludes moral criteria from the identification of the existence or content of valid, positive laws.

The interrelated questions asked in this Article are: (1) whether the concept of the rule of law is well served by a delimited focus on whether the legal system promulgates laws capable of being obeyed; (2) whether a morally richer view of the rule of law fits the concept’s use and formulation; and (3) whether, even if extending beyond traditional formulations, a rule of law inquiry broader than Hart’s capacity-to-obey test should be seen as conceptual overreach? The answers argued for here aspire toward both releasing the rule of law construct from its formal-equality fetters and accentuating the construct’s potential for improving the moral landscape endured by those for whom legal arrangements reinforce impairment of their capabilities to exercise powers that the legal system otherwise confers or encourages.

AI, Patent Reading and Patent Disclosure

'Misleading AI: Regulatory Strategies for Transparency in Information Intermediary Tools for Consumer Decision-Making' by Jeannie Marie Paterson in (2023) Loyola Consumer Law Review comments

Increasingly, consumers’ decisions about what to buy are mediated through digital tools promoted as using “AI”, “data” or “algorithms” to assist consumers in making decisions. These kinds of digital information intermediaries include such diverse technologies as recommender systems, comparison sites, virtual voice assistants, and chatbots. They are promoted as effective and efficient ways of assisting consumers making decisions in the face of otherwise insurmountable volumes of information. But such tools also hold the potential to mislead consumers, amongst other possible harms, including about their capacity, efficacy, and identity. Most consumer protection regimes contain broad and flexible prohibitions on misleading conduct that are, in principle, fit to tackle the harms of misleading AI in consumer tools. This article argues that, in practice, the challenge may lie in establishing that a contravention has occurred at all. The key characteristics that define AI informed consumer decision-making support tools ––opacity, adaptivity, scale, and personalization –– may make contraventions of the law hard to detect. The paper considers whether insights from proposed frameworks for ethical or responsible AI, which emphasise the value of transparency and explanations in data driven models, may be useful in supplementing consumer protection law in responding to concerns of misleading AI, as well as the role of regulators in making transparency initiatives effective.

'Linguistic metrics for patent disclosure: Evidence from university versus corporate patents' by Nancy Kong, Uwe Dulleck, Adam B Jaffe, Shupeng Sun and Sowmya Vajjala in (2023) 52(2) Research Policy comments 

 Encouraging disclosure is important for the patent system, yet the technical information in patent applications is often inadequate. We use algorithms from computational linguistics to quantify the effectiveness of disclosure in patent applications. Relying on the expectation that universities have more ability and incentive to disclose their inventions than corporations, we analyze 64 linguistic measures of patent applications, and show that university patents are more readable by 0.4 SD of a synthetic measure of readability. Results are robust to controlling for non-disclosure-related invention heterogeneity. The linguistic metrics are evaluated by a panel of “expert” student engineers and further examined by USPTO 112(a) – lack of disclosure – rejection. The ability to quantify disclosure opens new research paths and potentially facilitates improvement of disclosure. ... 

The patent system serves two purposes: “encouraging new inventions” and “adding knowledge to the public domain”. The former incentivizes creation, development, and commercialization by protecting inventors’ exclusive ownership for a limited period of time. The latter encourages disclosure of new technologies by requiring “full, clear, concise, and exact terms” in describing inventions.2 Sufficient disclosure in patents has three major benefits: (1) fostering later inventions (Jaffe and Trajtenberg, 2002, Scotchmer and Green, 1990, Denicolò and Franzoni, 2003); (2) reducing resources wasted on duplicate inventions (Hegde et al., 2022); and (3) inducing more informed investment in innovation (Roin, 2005). 

Despite a large body of literature on the patent incentivizing function (Cornelli and Schankerman, 1999, Kitch, 1977, Tauman and Weng, 2012, Cohen et al., 2002), patent disclosure receives limited attention. This raises concerns; as Roin (2005), Devlin (2009), Sampat (2018), Arinas (2012) and Ouellette (2011) document, the technical information contained in patent documents is often inadequate and unclear. Important questions, such as how to measure disclosure, potential incentives behind disclosure, heterogeneous levels of disclosure by entities, and the tactic of avoiding the disclosure requirement, have not been directly investigated. A major barrier to such empirical research has been the lack of broadly applicable, reproducible quantitative measures of the extent of disclosure or information accessibility. We propose and demonstrate that extant metrics developed in computational linguistics can help to fill this gap. 

In using computational linguistic metrics to compare the readability of documents, we follow researchers in the finance and accounting literature, who have used readability metrics to gauge whether readers are able to extract information efficiently from financial reports (Li, 2008, Miller, 2010, You and Zhang, 2009, Lawrence, 2013). This literature posits that more complex texts increase the information processing cost for investors (Grossman and Stiglitz, 1980, Bloomfield, 2002) and finds, for example, that companies are likely to hide negative performance in complicated text to obfuscate that information (You and Zhang, 2009). 

Although patent applications differ from corporate annual reports, the research question regarding strategic obfuscation is similar: Documents are created subject to regulation, in which the purpose of the regulation is to compel disclosure, but the party completing the document may have incentives to obscure information. Our proposed linguistic measures are likely to serve as an informative proxy for the explicitly or implicitly chosen level of disclosure. The goal of this article is simply to demonstrate that these measures do appear to capture meaningful differences in accessibility or disclosure, and thereby opening up the possibility of research on the causes and effects of variations in disclosure. 

Our strategy for demonstrating the relevance of linguistic readability metrics is to identify a situation in which we have a strong a priori expectation of a systematic difference in disclosure across two groups of patents. If the proposed metrics show the expected difference, we see this as an indication to treat them as potentially useful. We compare patent applications from universities with those of corporations. Both strategic reasons and the costs of revealing information inform our expectations. From a strategic perspective, universities, with their focus on licensing of patents have an interest in making their patents more accessible. In contrast, corporations (particularly practicing corporations) may benefit from limiting the accessibility of information. From a cost perspective, drafting patents is usually informed by documentation of the relevant research or process of innovation. Given university researchers’ primary interest in accessible publications and the relevant standards of documentation, the source material available to an attorney drafting a patent may be much better than in the case of the same attorney drafting a patent for a corporation, in which the need for such documentation is much less. The literature also supports this expectation (Trajtenberg et al., 1997, Henderson et al., 1998, Cockburn et al., 2002). 

Universities and corporations follow different business models for patenting: technology transfer versus in-house commercialization. Patents applied for by universities, with a focus on generating income from the licensing of inventions, should have a higher level of disclosure because transparent information makes it easier to signal the technology contained in the patent and attract potential investors. As a result, they are more readable than corporate patents. The readability difference could be further magnified by the moral requirements of university research as well as the rigor of academic writing, which could further affect the level of disclosure. 

Corporations, particularly those with a focus on in-house production, on the other hand, have a greater incentive to obfuscate crucial technical information to deter competitors from understanding, using, and building on their inventions. The profit-maximizing motive, as well as a lack of incentive to thoroughly document the invention, could also contribute to the low level of disclosure. Together, it is reasonable to assume that universities may strategically (or unconsciously) choose a higher disclosure level in patent applications than corporations. We emphasize that we do not see this analysis as testing the hypothesis that universities engage in more disclosure than corporations for a particular reason. Rather, we take this as a maintained hypothesis and show – conditional on that maintained hypothesis – that the linguistic measures meaningfully capture differences in disclosure across patents, which indicates the value of further research and the need to reconsider patent examination with respect to the accessibility and disclosure of information contained in patents. 

Similar to the finance literature, we use a computational linguistic program designed to assess the reading difficulty of texts using 64 measures from second language acquisition research. The indicators cover the lexical, syntactic, and discourse aspects of language along with traditional readability formulae. We apply them to a full set of U.S. patent application texts in three cutting-edge industries from the past 20 years. Our baseline OLS estimations reveal significant differences between university and corporate patents. Using principal component analysis (PCA) to combine the 64 indicators and create synthetic readability measures, we show that composite indices detect strong differences between university and corporate patents, which lends support to the validity of our measures. 

The key empirical challenge is that the nature of corporate and university inventions might differ; thus, the textual communication required for corporate inventions could differ. To address this concern, our identification strategy employs the following. First, to account for the unobserved heterogeneity in linguistic characteristics intrinsic to technical fields, our econometric method controls for U.S. patent subclass fixed effects. This enables us to measure disclosure as the degree of readability relative to other technologically similar patents. Second, we use patent attorney fixed effects to control for systematic disclosure effects from the drafting agents. This compares the university and corporate patents drafted by the same patent attorney. Third, we employ cited-patent fixed effects with a data compression technique, least absolute shrinkage and selection operator (LASSO), to further control for the nature of inventions. This is because university and corporate patents that cite the same previous patents build on the same prior knowledge, and are therefore likely to be technologically similar inventions. Fourth, to deal with any selection bias from observables, we use a doubly robust estimation that combines propensity score matching and regression adjustment. This enables us to compare university and corporate patents with similar attributes. 

Our results show that corporate patents are 0.4 SD more difficult to read and require 1.1–1.6 years more education to comprehend than university patents. We find that the difference is more prominent for more experienced patent applicants, and that licensing corporate patents disclose more than other corporate patents, which we believe supports the idea that the differences in readability are at least somewhat intentional. We also show that a potential channel for obfuscation lies in the provision of many examples in order to conceal the “best mode” of inventions. 

This paper is one of the first to specifically use textual analysis to examine patent disclosure (with exception of Dyer et al. (2020) who focus on patent examiners’ leniency) and validate the measure. We obtain the whole set of full text patent applications in categories related to nanotechnology, batteries, and electricity from 2000 to 2019, totaling 40,949, and apply our linguistic analysis model to the technical descriptions of these patents. We expand readability studies in related literature that rely heavily on traditional readability indices such as Gunning Fog, Kincaid, and Flesch Reading Ease by including lexical richness, syntactic complexity, and discourse features. We use the best non-commercial readability software (Vajjala and Meurers, 2014b) to capture the multidimensional linguistic features of 64 indicators, and perform a more in-depth linguistic analysis (Loughran and McDonald, 2016) than previous studies. We also use principal components analysis to construct synthetic overall measures of readability. 

Having developed this rich set of readability measures, we validate them as indicators of effective patent disclosure by testing whether the lexical measures show patents to be more readable in several real-world contexts. Our primary comparison is between university and corporate patents. The licensing aims of universities and absence of market driven competitive motives mean that they have greater incentive to disclose – less incentive to conceal – key information relative to corporations. Through analyses that control for sources of variation in readability, we find that university patents are, indeed, more readable. We support this main analysis with several other comparisons. Intellectual Ventures – a corporation that, akin to universities, seeks to license its patents over competing in the market – also holds patents with above average readability. Several large corporations known to be active patent licensors (IBM, Qualcomm, and HP) similarly exhibit higher readability. Additionally, a set of patents that can be presumed to have been reassigned also exhibit higher readability than otherwise similar patents. Finally, we compared the computational readability measures to subjective evaluations of readability and disclosure for a small number of patents, and assessed the readability of patents rejected by the USPTO for reasons that include failure to adequately disclose the technology. 

We see the role of this paper as analogous to Trajtenberg et al. (1997), who first introduced metrics of patent “importance”, “generality” and “originality” based on patent citation data. We imitate their strategy to test whether our proposed new measures reveal the contrast we expect between university and corporate patents, and argue that the finding – that they display the predicted pattern – can be taken as initial evidence that they capture meaningful variation in unobservable patent disclosure quality. The introduction and initial validation of these measures open up the possibility of quantitative treatment of extent of disclosure in patents, both for social science research on the sources and effects of better or worse disclosure, and potentially for use in more systematic treatment of the disclosure obligation in the patent examination process. 

The rest of the paper proceeds as follows. Section 3 explains the linguistic measures used in the study. In Section 2, we review the relevant literature and lay out our hypothesis of differences in disclosure between university and corporate patent applications. Section 4 presents our data and baseline estimation, followed by our main results in Section 5. We examine attorney fixed effects and cited-patent fixed effects in Section 6, and one channel that corporations could use to obscure patent applications in Section 7. We show heterogeneous effects in Section 8 and usefulness tests in Section 9, and conclude in Section 10.

Ecocentrism

'Implementing Rights of Nature: An EU Natureship to Address Anthropocentrism in Environmental Law' by Niels Hoek, Ivar Kaststeen, Silke van Gils, Eline Janssen, and Marit van Gils in (2023) 19(1) Utrecht Law Review 72–86 comments 

Transboundary issues – from (chemical) pollution, land-use change to unsustainable levels of exploitation – have been eroding natural sites across Europe, reducing biodiversity in the process. In light of this, this paper analyses the comprehensiveness of EU environmental law, appraising its underlying ethos in the process. Additionally, it explores whether a Natureship Framework Directive at the European Union (EU) level, which establishes legal personality for natural sites, can deliver a ‘change of course’ with respect to the anthropocentric view underpinning environmental law as a pressing thought experiment. It constructs a (fictive) law which grants natural sites substantive and procedural rights, conceptualising how such an instrument may take shape. One finding is that an EU Natureship may be a robust tool to address flaws within EU environmental law. For example, the attribution of legal personality to natural sites alongside the appointment of formal representatives can significantly relieve the burden for NGOs and the European Commission, which may suffer from limited resources when it comes to judicial enforcement of environmental norms (or, alternatively, ecological rights). Other benefits pertain to nature management, which may be less complex and more politically stable under the approach put forward in this paper. An EU Natureship, therefore, may provide a vehicle to shift EU environmental law from the anthropocentric to the ecocentric.

The authors consider 'Rights of Nature (RoN) as a response to failing protection', stating 

The degradation of natural ecosystems continues, as confirmed by the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services (IPBES), despite the adoption of a wide degree of national, regional, and international nature conservation instruments. Against this backdrop of continuing ecological decay, new forms of protection have been proposed by international lawyers, NGOs, and academics, which challenge existing environmental laws. One such regulatory approach is the assignment of legal personhood to natural sites, which directly grants standing in court and, perhaps more pressingly, confers substantive and procedural rights to said natural sites. In a general sense, the rights of nature movement mirrors how the law confers rights (but not obligations) to individuals, companies, or institutions. This controversial yet much-cited idea was first put forward by Christopher Stone in 1972. Fundamentally, it represents a shift from a view of nature as an object before the law to a view of nature as a subject of the law. This movement has gained prominence, given the pressures which natural ecosystems face. A practical case of ecological decline can be found in the Dutch, German and Danish Wadden Sea, where climate change, pollution, and large-scale mining activities are causing irreversible damage to the natural site, with modern-day legal instruments seemingly unable to halt this decline. For example, in 2022, a permit for mining was issued in the Netherlands, approving the further exploitation of the Wadden Sea, despite a backlash from a plethora of NGOs and local residents. 

In light of these pressures, the case for assigning legal personhood to the Wadden Sea was made by Lambooy and others in 2019. In their article, the authors argued in favour of adopting legal personhood for the Dutch part of the Wadden Sea. In this context, they put forward the idea of a ‘Natureship’. Linguistically speaking, a Natureship places the focus on the underlying entity, namely, the natural site. The suffix -ship implies a position held and/or created, a grammatical feature common in both Dutch and English. Their article defines a Natureship as a ‘public law person’ that combines the power of a public institution, such as environmental management, with private powers, such as the ability to own assets or claim reparations. The statutory purpose of the Natureship would be to ‘protect and support the ecological coherence’ of a specified geographical area, with significant independence from external governmental interference. Lambooy and others argue that, under Dutch law, legal personhood can be granted to natural sites in this form. Here lies the relevance of this contribution: the concept of a Natureship implements the Rights of Nature movement in practical terms. The idea has received traction in the Netherlands, where the ‘rights for the Wadden Sea’ has been transformed from a foreign concept into an issue seriously contemplated within the national Parliament. 

However, while this paper by Lambooy and others is a highly valuable contribution to legal scholarship, it does not take into account the transboundary nature of most ecosystems. The Wadden Sea, for example, spans three different EU Member States, namely, the Netherlands, Germany, and Denmark. Biodiversity does not inherently subscribe to the idea of human-made borders; thus, a Natureship grounded in EU law is a proposition that needs further exploration. An EU approach may tackle transboundary issues more effectively than a purely national one since the latter cannot reach the desired spatial scope nor guarantee uniform protection throughout all the concerned Member States. This paper will explore the merits of an EU approach to Natureships by engaging in a thought experiment, conceptualising how such an instrument may take shape at the EU level. It refers to the German, Dutch and Danish Wadden Sea as an example to reflect on the potential merits of this approach, when needed. 

Overall, this will inform the question of whether an EU Natureship law may act as a ‘remedy’ for anthropocentrism within EU environmental law. As an ethos underpinning the law, anthropocentrism takes a human-centred approach to legislation. This is the antithesis of an EU Natureship. In contrast, an ecocentric ethos subscribes intrinsic value to nature as a collective. An ecocentric law, as such, provides a holistic perspective towards environmental protection, including non-human interests within the scope of consideration. An EU Natureship, in essence, is a proposition aimed to achieve such holistic protection. 

This paper hosts several methodologies to unpack the research statement as outlined above. The primary method deployed consists of doctrinal research, initially taking an ‘internal’ perspective of the legal system. However, from the normative premise that the law may be failing the environment, a critical analysis will be undertaken, exploring the possible routes ahead in order to close the perceived gap between law and ecology. This article will highlight anthropocentrism within modern-day EU environmental law in Section 2.1 and 2.2, thus analysing its failures. In the following Sections 3.1, 3.2 3.3 and 3.4, the concept of an EU Natureship is set out as a thought experiment, taking the Wadden Sea wetland as an example. In doing so, concrete provisions are suggested in the context of a (fictive) EU Natureship Regulation and/or Framework Directive. The paper ends with a brief conclusion, in Section 4, on the merits of such an EU-wide approach. It should be noted that this paper does not review international obligations derived from the Ramsar Convention, the UNESCO World Heritage Convention, or the Convention on Biological Diversity. Whilst international instruments are essential components of the legal framework that governs wetland protection, given the EU-specific proposition being put forward, the scope of this paper will primarily be limited to the supranational.

FDA

'The Fall of FDA Review' by Daniel G. Aaron in (2023) Yale Journal of Health Policy, Law, and Ethics comments 

The U.S. Food and Drug Administration is in crisis. Contaminated baby formula is only the latest of a series of scandals that have catapulted FDA into the public spotlight. FDA can hardly go a single day without an investigation, news scoop, or scholarly critique of the agency’s work. FDA regulates 25% of the U.S. economy, yet the array of problems facing the agency raises questions about whether it is equipped to succeed in the 21st century. 

FDA’s core function is to oversee a special legal regime called “premarket review.” Congress has prohibited all marketing of certain types of products (like drugs) until FDA reviews and approves an application from the manufacturer. This system allows consumers to depend on the foods they ingest, the pills they swallow, the health care they receive—in theory. But critics have documented how FDA review failures have produced, or contributed to, public health crises, including those related to opioids, e-cigarettes, trans fats, sugar, and, most recently, the COVID-19 pandemic. U.S. life expectancy fell by 2 years between 2018 and 2020, in part due to FDA-regulated products. What can explain this extreme abdication of regulatory authority that is leaving the American public unsafe and unprotected? 

Until now, the siloed nature of FDA law has prevented a meaningful analysis of premarket review as a legal regime. This article offers the first cross-disciplinary critique of premarket review across five FDA product areas. Leveraging regulatory history, medical science, epidemiology, and law, I argue that premarket review is faltering and aim to explain why. The reasons vary somewhat across FDA’s regulatory regimes. However, the bottom line is the same: Longstanding efforts to undermine FDA governance by corporations and financial power writ large. Corporate deregulatory efforts have operated through courts, Congress, the President, and the agency’s leadership itself. In some cases, premarket review has been so hollowed out that all that remains is the illusion of regulation, nothing more. These developments reflect the ascendancy of neoliberalism, a system in which core social guarantees devolve to decisions by individual consumers. 

We need not accept this state of affairs. Learning from the mechanisms behind premarket review’s erosion, I propose a suite of structural solutions to build a revitalized FDA: one that is dutifully empowered, inside and out, to safeguard the public health.

Biometrics

'Suspect AI: Vibraimage, Emotion Recognition Technology and Algorithmic Opacity' by James Wright in (2021) Science, Technology and Society comments 

 Vibraimage is a digital system that quantifies a subject’s mental and emotional state by analysing video footage of the movements of their head. Vibraimage is used by police, nuclear power station operators, airport security and psychiatrists in Russia, China, Japan and South Korea, and has been deployed at two Olympic Games, a FIFA World Cup and a G7 Summit. Yet there is no reliable empirical evidence for its efficacy; indeed, many claims made about its effects seem unprovable. What exactly does vibraimage measure and how has it acquired the power to penetrate the highest profile and most sensitive security infrastructure across Russia and Asia? xx I first trace the development of the emotion recognition industry, before examining attempts by vibraimage’s developers and affiliates scientifically to legitimate the technology, concluding that the disciplining power and corporate value of vibraimage are generated through its very opacity, in contrast to increasing demands across the social sciences for transparency. I propose the term ‘suspect artificial intelligence (AI)’ to describe the growing number of systems like vibraimage that algorithmically classify suspects/non-suspects, yet are themselves deeply suspect. Popularising this term may help resist such technologies’ reductivist approaches to ‘reading’—and exerting authority over—emotion, intentionality and agency.

Wright states 

As I sat in the meeting room of a nondescript office building in Tokyo, the managing director of a company called ELSYS Japan discussed my emotional and psychological state, referring to a series of charts and tables displayed on a large screen at the front of the room: Aggression … 20-50 is the normal range, but you scored 52.4 … this is a bit too high. Probably you yourself didn’t know this, but you’re a very aggressive person, potentially… Next is stress. Your stress is 29.2, within the range of 20-40, with a statistical deviation of 14—that’s OK… I think you have very good stress… Just tension—your [average] value is within the range, but because your statistical deviation is high—over 20—so you’re a little tense. Mental balance is 64 from a range of 50-100, so it fits correctly in the range… Charm … 74.6 is pretty good. Now, neuroticism is 35.3, this is also in the range, but the statistical deviation is high. But some people have a high score the first time they are measured. There are people who have high scores for neuroticism as well as for tension, yes. People who possess a delicate heart.1 (Interview, 17 April 2019) xx The director’s seemingly authoritative statements were based on an assessment of various measurements produced by ‘vibraimage’, a patented2 system developed to quantify a subject’s mental and emotional state through an automated analysis of video footage of the physical movements of their face and head. This system, distributed in Japan by ELSYS Japan under the brands ‘Mental Checker’ and ‘Defender-X’, provides numerical values for levels of aggression, tension, balance, energy, inhibition, stress, suspiciousness,3 charm, self-regulation, neuroticism, extroversion and stability, categorising these automatically into positive and negative ‘emotions’. Mental Checker generates an impressive array of statistical data arranged across tables, pie chart, histogram and line chart, producing an image of mathematical precision and solid scientific legitimacy (see Figure 1). The report also provides a visualisation of what ELSYS Japan terms an ‘aura’—a horizontal colour-coded bar chart, indicating the frequency of micro-vibrations of a subject’s head, superimposed against a still image of their face. 

Vibraimage technology has already entered the global security marketplace. It was deployed at the 2014 Sochi Olympics (Herszenhorn, 2014), 2018 PyeongChang Winter Olympics, 2018 FIFA World Cup in Russia and at major Russian airports to detect suspect individuals among crowds (JETRO, 2019). It has been used at the Russian State Atomic Energy Corporation in experiments to monitor the professionalism of workers handling and disposing of spent nuclear fuel and radioactive waste (Bobrov et al., 2019; Shchelkanova et al., 2019), and to diagnose their psychosomatic illnesses (Novikova et al., 2019). In Japan, Mental Checker and Defender-X have been used by one of the largest technology and electronics companies, NEC,4 to vet staff at nuclear power stations and by a leading security services firm, ALSOK, to detect and potentially deny entry to or detain suspicious individuals at major events, including the G7 Summit in 2016, as well as sporting events and theme parks (Interview with ELSYS Japan, 17 April 2019). Managers at ELSYS Japan expected that the technology would be used at the 2020 Tokyo Olympics (Nonaka 2018, p. 148, Interview with ELSYS Japan, 17 April 2019), an event that spurred significant increased spending on domestic security services and infrastructure, with estimated market growth of 18% between 2016 and 2019 (Teraoka, 2018).5 ELSYS Japan’s customers also include Fujitsu and Toshiba, which have considered ‘incorporat[ing] [vibraimage]… into their own recognition technologies to differentiate their original products’ (Nonaka, 2018, p. 147), and managers told me that Mental Checker has been used by an unspecified number of Japanese psychiatrists to confirm diagnoses of depression. 

In South Korea, the Korean National Police Agency, Seoul Metropolitan Policy Agency and several universities have collaborated on research aiming to establish the use of vibraimage in a video-based ‘contactless’ lie-detection system as an alternative to polygraph testing (Lee & Choi, 2018; Lee et al., 2018), while, in China, it has been deployed in Inner Mongolia, Zhejiang and elsewhere to identify suspects for questioning and detention, and has been officially certified for use by Chinese police (Choi et al., 2018a, 2018b).6 Other corporate applications of vibraimage are also proposed: an ELSYS Japan brochure suggests using Mental Checker to discover how employees really feel about their company; measure their levels of stress, fatigue and ‘potential ability’; counter employees’ accusations of bullying and abuses of power in the workplace; and even ‘to know the risk of hiring persons who might commit a crime’ (ELSYS Japan Brochure, undated). The brochure provides a screenshot of a suggested employee report, with grades (A+, B−, C, etc.) for qualities that include stability, fulfilment and happiness, social skills, teamwork, communication, ability to take action, aggressiveness, stress tolerance and ability to ‘recognise reality’. 

Vibraimage forms one part of the rapid growth in algorithmic security, surveillance, predictive policing and smart city infrastructure across urban East Asia, enabling the ‘active sorting, identification, prioritization and tracking of bodies, behaviours and characteristics of subject populations on a continuous, real-time basis’ (Graham & Wood, 2003, p. 228). Amid an international boom in both surveillance technologies and artificial intelligence (AI) systems designed to extract maximal information from digital photographic and video data relating to the body, companies are developing algorithms that move beyond facial recognition intended to identify individuals and increasingly aim to analyse their behaviour and emotional states (AI Now Institute, 2018, pp. 50–52). The digital emotion recognition industry was worth up to US$12 billion in 2018, and it continues to grow rapidly (AI Now Institute, 2018). 

As the concepts of algorithmic regulation and governance (Goldstein et al., 2013; Introna, 2016) are increasingly becoming a reality, transparency has become a key theme in critiques of black-boxed algorithms and AI, including those used in emotion recognition. This is particularly the case with machine learning, in which algorithms recursively adjust themselves and can quickly become inexplicable even to data science experts. As Maclure puts it, ‘we are delegating tasks and decisions that directly affect the rights, opportunities and wellbeing of humans to opaque systems which cannot explain and justify their outcomes’ (Maclure, 2019, p. 3). Transparency is linked to and overlaps with values of comprehensibility, explicability, accountability and social justice, and it is frequently presented as a vital component of ethical or ‘good’ AI (Floridi et al., 2018; Hayes et al., 2020; Leslie, 2019). ... 

...  This article uses the case of vibraimage to examine issues around opacity and the work it does for companies and governments in the provision of security services, by attempting to shed light on the algorithms of vibraimage and its imagined and actual uses, as far as possible based on publicly available data. What exactly does vibraimage measure and how does the data the system produces, processed through an algorithmic black box, deliver reports that have acquired the power to penetrate corporate and public security systems involved in the highest profile and most sensitive security tasks in Russia, Japan, China and elsewhere? The first section of the article examines emotion detection techniques and their digitalisation. The second section focuses on vibraimage and how its proponents, many of whom have commercial relationships with companies distributing it, have engaged in processes of scientific legitimation of the technology while making claims for its actual and potential uses. The final section considers how the disciplining power and corporate value of vibraimage are generated through its very opacity, in stark contrast to increasingly urgent demands across the social sciences and society, more broadly, for transparency as a prerequisite for ‘good AI’. I propose the term ‘suspect AI’ reflexively to describe the increasing number of algorithmic systems, such as vibraimage, in operation globally across law enforcement and security services, which automatically classify subjects as suspects or non-suspects. Popularising this term may be one way to resist such reductivist approaches to reading and exerting authority over human emotion, intentionality, behaviour and agency. 

Emotion Recognition Based on Facial Expressions 

Psychologist Paul Ekman pioneered research exploring the relationship between emotions and facial expressions since the 1960s, building on Darwin’s (2012[1872]) work on evolutionary connections between the two among animals, including humans. Ekman conducted experiments around the world, aiming to demonstrate the universality of a handful of basic emotions (such as anger, contempt, disgust, fear, happiness, sadness and surprise) across all cultures and societies, and of their articulation through similar facial expressions (Ekman, 1992). This work was highly influential because it seemed to provide overwhelming empirical evidence that individuals of all cultures were able to ‘correctly’ categorise the expressions of people of their own and other cultures provided in photos, matching them to the ‘basic emotions’ they supposedly expressed (Ekman & Friesen, 1971).  

Ekman further argued that facial expressions could be used to identify incongruities between professed and ‘real’ emotions, enabling facial expression analysis to be used for lie detection (Ekman & Friesen, 1969). This attracted substantial interest from corporations concerned with ensuring the honesty of employees or gaining covert insights in business negotiations, and from governments and security forces concerned with identifying dissimulating and suspect individuals. Ekman and collaborators in this field like David Matsumoto formed companies, running workshops and holding consultations with corporations and public bodies about how to read subjects’ facial micro-expressions and behavioural cues to evaluate personality, truthfulness and potential danger. In 2001, the American Psychological Association named Ekman one of the most influential psychologists of the twentieth century (APA, 2002). 

The identification of emotions through facial expressions underwent digitalisation via machine learning techniques pioneered since the mid-1990s by Rosalind Picard and Rana el Kaliouby at Massachusetts Institute of Technology (MIT). They commercialised this new field of ‘affective computing’ via their venture capital–backed company Affectiva, founded in 2009, which provides emotional analysis software to businesses based on algorithms trained on large databases of facial expressions (Johnson, 2019). According to Affectiva, this enables a test subject’s emotional responses to, for example, TV commercials, to be tracked in real time. With the recent boom in facial recognition technology, emotion recognition represents a rapidly expanding area of AI development, used across industries, including recruitment and marketing research (Devlin, 2020). A growing number of companies offer emotion recognition services based on analysis of facial expressions, including Microsoft (Emotion application programming Interface [API]), Amazon (Rekognition), Apple (Emotient, which Ekman advised on) and Google (Cloud Vision API). 

Such systems are increasingly being used in border protection and law enforcement to identify dissimulating and otherwise suspect individuals, regardless of substantial evidence of efficacy. From 2007, the Transportation Security Administration (TSA) spent US$900 million on a ‘behaviour-detection programme’ entitled Screening Passengers by Observation Technique (SPOT), until it was ruled ineffective by the Department of Homeland Security and the Government Accountability Office (GAO, 2013). Ekman consulted on SPOT, and the system incorporated his techniques; his company also provided consulting services to US courts (Fischer, 2013). Another system—Automated Virtual Agent for Truth Assessments in Real-Time (AVATAR), was developed for lie detection targeting migrants on the USA–Mexico border (Daniels, 2018), while the EU trialled the iBorderCtrl system, supplied by the consortium European Dynamics and funded by Horizon 2020, using the interpretation of micro-expressions to detect deceit among migrants in Hungary, Greece, and Latvia (Boffey, 2018; see also AI Now Institute, 2018, pp. 50–52). 

Recently, this work on facial expression analysis for emotion recognition has come under increasing scrutiny despite its ongoing popularity among many psychologists. The most basic critique is that one does not necessarily smile when one is happy—common sense suggests that facial expressions do not always, or even often, map to inner feelings, that emotions are often fleeting or momentary, and that facial expressions and their meaning are highly dependent on sociocultural context. Barrett et al. (2019) summarise these and other critiques, arguing that approaches positing a limited number of prototypical basic emotions that can be ‘read’ through universal facial expressions fail to grasp what emotions are and what facial expressions convey. 

In anthropology, the ‘affective turn’ has drawn attention to the distinction between affect and emotion—the former a precognitive sensory response or potential to affect and be affected, and the latter a more culturally mediated expression of feeling. White describes this as the difference between ‘how bodies feel and how subjects make sense of how they feel’ (White, 2017, p. 177). These nuances are overlooked in the field of emotion recognition, which reduces emotion to a simplistic and digitally scalable model. Barrett argues that emotion is: a contingent act of perception that makes sense of the information coming in from the world around you, how your body is feeling in the moment, and everything you’ve ever been taught to understand as emotion. Culture to culture, person to person even, it’s never quite the same. (Fischer, 2013) 

We might, therefore, define the process of interpreting one’s own emotional state as making sense of an inner noise of biological signals and memories, in contextually contingent and socioculturally mediated ways, and placing them into—and in the process co-constructing—socioculturally mediated categories. It may also sometimes involve not definitively categorising or making sense of these affective feelings. As this article will show, it is the very ambiguity or malleability of this process that may help make vibraimage a convincing technology of emotion recognition and provide authority to its analysis. 

Given these growing critiques of Ekmanian theories of universal basic emotions expressed through facial expressions, researchers at the organisation AI Nowhave concluded that, by extension, the digital emotion detection industry is ‘built on markedly shaky foundations…. There remains little to no evidence that these new affect-recognition products have any scientific validity’ (AI Now Institute, 2018, p. 50). Baesler, similarly, argues that the use of emotion detection software by the TSA was ‘unconfirmed by peer-reviewed research and untested in the field’ (Baesler, 2015, pp. 60–61), while holding significant potential for harm through misuse. In common with broader critiques of AI from critical algorithm studies (e.g., Eubanks, 2018; Lum & Isaac, 2016), machine learning methods involved in emotion recognition systems have been criticised for racial bias, based on their training data sets (Rhue, 2018). Indeed, Ekman’s work not only constructs ethnocentric emotional categories but also racial subject categories, for example in his creation, with Matsumoto, of the Japanese and Caucasian Facial Expressions of Emotion stimulus set of photos showing emotional expressions of archetypal ‘Japanese’ and ‘Caucasian’ subjects (Biehl et al., 1997; https://www/humintell.com), which continues to be used in psychology experiments. For all of these reasons, the increasingly widespread application of this technology has raised growing ethical and civil liberties concerns

'Automated Video Interviewing as the New Phrenology' by Ifeoma Ajunwa in (2022) 36 Berkeley Technology Law Journal 101 comments 

This Article deploys the new business practice of automated video interviewing as a case study to illuminate the limitations of traditional employment antidiscrimination laws. Employment antidiscrimination laws are inadequate to address unlawful discrimination attributable to emerging workplace technologies that gatekeep equal opportunity in employment. The Article shows how the practice of automated video interviewing is based on shaky or non-proven technological principles that disproportionately impact racial minorities. In this way, the practice of automated video interviewing is analogous to the pseudo-science of phrenology, which enabled societal and economic exclusion through the legitimization of eugenics and racist attitudes. After parsing the limitations of traditional anti-discrimination law to curtail emerging workplace technologies such as video interviewing, this Article argues that ex ante legal regulations, such as those derived from the late Professor Joel Reidenberg’s Lex Informatica framework, may be more effective than ex post remedies derived from the traditional employment antidiscrimination law regime. The Article argues that one major benefit of applying a Lex Informatica framework to video interviewing is developing legislation that considers the capabilities of the technology itself rather than how actors intend to use it. In the case of automated hiring, such an approach would mean actively using the Uniform Guideline on Employee Selection Procedures to govern the design of automated hiring systems. For example, the guidelines could dictate design features for the collection of personal information and treatment of content. Other frameworks, such as Professor Pamela Samuelson’s “privacy as trade secrecy” approach could govern design features for how information from automated video interviewing systems may be transported and shared. Rather than reifying techno solutionism, a focus on the technological capabilities of automated decision-making systems offers the opportunity for regulation to start at inception, which in turn could affect the development and design of the technology. This is a preemptive approach that sets standards for how the technology will be used and is a more proactive legal approach than merely addressing the negative consequences of the technology after they have occurred.