Building trust in the digital era: achieving Scotland's aspirations as an ethical digital nation: case study supplement

This paper is a supplement to the ‘Building Trust in the Digital Era: Achieving Scotland’s Aspirations as an Ethical Digital Nation’ Report. The case studies have fed into the core report content, helping to position the ethical challenges relating to digital innovation across a range of sectors.


The Future of Work in a Digital Economy

Case Study: Governing the Rise in the Remote Economy – Sam Brakarsh & Abigail Marks

The nature of the relationship between home and work has changed significantly over the past 18 months. The Office of National Statistics (ONS) noted that prior to the start of 2020, employees that worked from home were most likely to be female and work part-time hours. Yet in 2020, given the unprecedented increase in homeworking, the characteristics of the home working population changed. Full-time, ICT, and professional workers now represent the highest incidence of homeworking. Despite some evidence suggesting that, people who combine homeworking with working away from home are more productive than those who never work from home, Scotland both during and before the pandemic had the lowest proportion of home and hybrid workers (ONS, July 2020).

Boss Tech and Surveillance Tech programs are playing an increasingly central role in the experience of employment and the structure of the labour market. Currently, industry and government policies are promoting remote work as a means to address inequality by increasing productivity and access to opportunity regardless of geographic location. However, an unreserved emphasis on remote work has the potential to cause significant social and economic harm. This case outlines three key domains in which remote work can fail to lead to social progress. Firstly, the language of inequality reduction can be co-opted by the for-profit industry to reduce costs to the detriment of community wellbeing. Secondly, the expansion of remote work only increases access to opportunity for a stratum of the labour force already engaged in work that is easily digitised. Indeed, the TUC has warned of an “emerging class divide” between higher-paid homeworkers who plan to continue doing their job remotely at least some of the time, and working-class occupations where people have limited access to, or opportunity for, any form of flexible working. Finally, remote work has the potential to significantly increase gender disparities in the workplace.

The Digital Strategy for Scotland sets as a goal that the country becomes a centre for home working, saying “we will engage with communities in remote and rural areas to find ways in which Scotland can capitalise on changes in the world of work and position itself as a leading centre for home and remote working.” The articulated value of such an initiative is clear. It has the potential to increase the ease of work for entrepreneurs who would benefit from collaborations that extend beyond the local. Boss Tech also allows teams to coordinate without shared office spaces thereby purportedly increasing worker efficiency without in situ managerial supervision. In addition, remote work is framed as a solution to spatial inequality, opening up access to specialised employment for individuals in remote areas of Scotland who would otherwise find such opportunities restricted to those in larger cities and business centres. The potential for these benefits holds true. However, an overemphasis on remote work can lead to social harm. Comprehensive policy is needed to secure the economic and social gains whilst protecting against potential drivers of inequity that are embedded within remote work.

The language of inequality reduction can be co-opted by the for-profit industry to justify actions that may benefit the employer rather than the employee. Remote work can increase access to opportunity, but remote work is just as capable of being used as a strategy to shift costs of supplies and overheads onto employees. Throughout the COVID-19 pandemic, large organisations have provided insufficient support to employees to work from home. Most individuals have had to pay for basic office supplies out of pocket to make their home environment workable.

At least a quarter of employees had to finance provision for IT tools in order to homework during the COVID-19 pandemic and over half had to provide their own office equipment. In addition, working remotely is likely to silo employees and make collective mobilisation around worker rights more challenging. Shared hardships in the workspace are veiled through restricted social engagements on digital platforms. As one participant from the Working@Home project noted ‘And with the technology, you know, you can see people, talk to people…..I think you missed some of the contact with people you’re particularly friendly with ... So, there’s probably been occasions where it would be nice if, you know, we could get together’. Asking the question of who remote work serves, the employee or the corporation is vital in assessing how to protect the wellbeing and rights of Scottish peoples in the digital economy.

Remote work has its own barriers to entry. Blue-collar workers, whose labour is inextricably tied to their bodies, or those who work in Scotland’s extensive tourist industry, will not benefit from policies aimed at increasing remote work. From a recent TUC (2021) survey of employers, there is the suggestion that organisations are less likely to offer flexible work to staff who were unable to work from home during the pandemic. One in six (16 percent) of employers surveyed said that after the pandemic, they will not offer flexible working opportunities to staff who could not work from home during the pandemic, compared to one in sixteen (6 percent) saying they will not offer flexible working opportunities to those who worked from home during the pandemic.

We cannot allow flexible working to become a perk for the favoured few - offered to a minority of the workforce who are able to work from home – and serving to reinforce existing inequalities. The remote economy is most likely to support middle to upper-income workers whose skills are easily transferable to digital platforms. Remote work is not radically restructuring the economy. It removes some barriers to entry but introduces others. In particular, the Working@Home project found that those who had large homes with less occupancy (and thus the space to afford a dedicated office) were more likely to ‘succeed’ at homeworking. From the Working@Home survey, it was clear that this advantaged men with 60% of men having a dedicated home office space compared to 49% of women. Moreover, with many remote workers having to pay for some of their own office equipment and IT provision, there is another advantage to those that are most affluent. The impact of a crowded home on homeworking is illustrated in the following quote -

“I would say my home is no longer – well it is my home but it is very difficult to say office workspace and private are separate because of the setting. When I go into the kitchen, I see my laptop, I see the computer, or when I go into the living room I see the office, I see my workspace. So I am – ideally I’d have a room that I can lock up and close the door and that’s it but it is, then my work is constantly with me because I see the desk, I see the chair so I see whenever I move in the flat -” (male, hospitality)

Finally, current data indicates that remote work disproportionately harms women in the workforce. Working from home is frequently presented as a means for women to fill the conventionally gendered role of raising children whilst simultaneously maintaining their careers. Whilst the Working@Home data suggests that there is no clear difference in terms of men and women and their intention/desire, with all employees wishing to work at home around 50% of the time, this is a simplistic finding. The study clearly found that men are less distracted when working from home - compared to the office - and women are more distracted. Women’s reaction is probably due to their disproportionate responsibility for domestic chores and decreased opportunity for a dedicated office. Nonetheless, the majority of workers felt more productive at home.

This increase in productivity comes at a cost. From a survey (n=1200) undertaken in December 2020/January 2021 by the working@home project, workers were starting to suffer as a response to the remote working technology. Respondents report a clear relationship between the time they spend using collaborative or video conferencing software and the intensity of their overall work demands. Specifically, the respondents reported that the more time they spent using collaborative or video conferencing software, the greater their perception of ‘always’ working and the more they felt they were too frazzled to participate in family activities/responsibilities. One third (32%) of respondents believed that technology forces them to work to increasingly tight time schedules; more than one quarter (27%) of respondents reported that technology forced them to work more than they can handle; and one quarter (25%) of respondents reported that technology forces them to work faster than they were used to.

The Working@Home study was undertaken with employees across the employment spectrum. However, research focussing only on senior managers has indicated that men are far more interested in returning to the office. This difference has serious implications for gender equity in the workplace since all-male ‘managerial’ workspaces will become increasingly inaccessible to women and, in the case of promotions, having a physical presence in the office may increase the visibility and the chances of moving up the organisational ladder. There is reason to be concerned that mismanaging the rise in remote work could push gender equity in the workplace back over a decade.

Recommendations

  • Flexible/remote or hybrid work has to be encouraged across the board from senior to entry-level staff, regardless of gender, to avoid disparity
  • Organisations should be legally mandated to ensure equality in terms of promotion and progression regardless of work location
  • With the likely savings from the reduction in real estate and utilities costs, employers must financially support hybrid and home working to avoid penalties to the most vulnerable
  • Organisations must ensure that home and hybrid workers have appropriate, subsidised equipment at home
  • Employers must ensure appropriate integration and inclusion of homeworkers in organisational decisions making
  • Government must mandate that surveillance technology, whether in built-in software such as Teams, Zoom, or other programs more integral to performance systems, must only be used with the full consent and understanding of employers and employees

Case Study: Non-Fungible Tokens (NFTs) and Cultural Heritage – Dr. Foteini Valeonti & Prof . Melissa Terras

Non-Fungible Tokens (NFTs) are a new, blockchain-based technology that introduces scarcity into the digital realm. First invented in 2017 as a way for game players to trade virtual goods, NFTs emerged into the mainstream in 2021, when the NFT titled “5000 Everydays: The first 5000 days” featuring a collage by digital artist Mike Winkelmann was auctioned at Christie’s for 69 million US dollars (Kinsella 2021). The year before, 2020, was a particularly challenging year for museums and galleries, as, due to the pandemic, institutions had to fight for their financial survival with widespread redundancies and financially-motivated deaccessions (Valeonti et al. 2021). Museums Galleries Scotland were urging the government in August of 2020 to provide additional financial support to the sector stating that two thirds (71%) of the country’s independent museums had not enough funds to survive for one year (Knott 2020). Emerging in that context, NFTs were deemed as a potential lifeline for museums and galleries (Ciecko 2021). Major museums were amongst the early adopters of this new technology, raising funds by selling primarily digitised images of artefacts in their collections raising issues of digital deaccessioning (Valeonti 2022). The first to do so was the Uffizi Gallery, Italy, which sold an NFT on an image of Michelangelo’s Doni Tondo for 170,000 US dollars in June 2021 (Artnet News 2021); the Hermitage Museum in Russia sold 5 works for 440,000 US dollars (Partz 2021) and in the fall the British Museum sold multiple editions of NFTs of Hokusai woodblock prints following up with NFTs of paintings by J.M.W. Turner (Harris 2022).

Scotland is spearheading research in the broader field of blockchain technology. The University of Edinburgh’s Blockchain Technology Laboratory was co-established by IOHK (the parent company of the Cardano blockchain network) and since 2017, it is undertaking all pioneering research and development of Cardano, currently the 4th largest blockchain with a market capitalisation of 16 billion US dollars (CoinMarketCap 2022). In the cultural sector, numerous artists in Scotland have been “minting” (i.e. creating) and successfully selling NFTs, including graduate of the Edinburgh College of Art and Edinburgh-based artist Trevor Jones, who is amongst the top selling NFT artists of all time (Anon. 2022a). As of yet, no Gallery, Library, Archive or Museum in Scotland has been known to mint and publicly auction an NFT as a means to raise funds. Although the potential of NFTs is indeed substantial, presenting a modern revenue stream, for the modern artworld and the increasingly digital cultural heritage sector, there are numerous issues to be considered in this emerging area, including understanding the environmental impact of NFTs and the underlying blockchain technology; understanding how copyright and intellectual property rights intersect with NFT ownership; navigating malicious uses of this technology and understanding the financial risks associated with investing in NFT artworks. However, these are issues that can be addressed, and since this is a rapidly emerging area that Scotland has shown leadership in, support should be provided for creators and institutions alike to understand and leverage this powerful new technology.

NFTs derive from the blockchain, a technology that enables trust-less transactions between strangers, eliminating the need for the mediation of centralised authorities; in simple terms, the blockchain can be described as a decentralised database that is accessible by everyone, but cannot be compromised by anyone (Valeonti et al. 2021). Beyond underpinning crypto-currency, blockchain technology is used for a wide range of applications (Ahram et al. 2017) considered also as the technological foundation of “web3” (Potts & Rennie 2019); The “decentralised web” as it is also referred to, web3 is widely considered the next iteration of the internet, as its decentralised nature could help tackle critical limitations of current Internet architecture (Jain 2006). By definition, NFTs are defined as cryptographically-unique, indivisible, irreplaceable and verifiable tokens that represent a given asset be it digital, or physical, on the blockchain (Valeonti et al. 2021). NFTs can be described as the digital equivalent of Limited Editions, which is a method widely employed by photographers and artists of immaterial art in order to create artificial scarcity due to the lack of a unique, original, tangible artefact. As demonstrated in Figure 1, in NFTs, the print copy is substituted with a digital copy and the signature is digital instead. Although NFTs closely resemble Limited Editions in many respects, the fundamental difference between the two is that, unlike Limited Editions, NFTs can compete in price with rare original artworks. Beeple’s “Everydays: The First 5000 Days” ranked at the time, as the 3rd most expensive work sold at an auction by a living artist (Kinsella 2021). As such, NFTs significantly increased the value of immaterial art (e.g. digital art, photography, performance), as well as the value of any kind of high quality, born-digital content.

With the trading volume of NFTs estimated at 40 Billion US dollars for 2021 (Versprille 2021), the potential of this new medium for fundraising is evident. Critically, despite the turbulent bear market, which is becoming increasingly severe, with all crypto-currencies currently in freefall, in a 7-day period (between the 12th and the 17th of June 2022), more than 18 million US dollars were invested in NFTs of Bored Apes alone, with 1 Bored Ape being sold for 1.22 million US dollars (Anon. 2022b). In the heart of the so-called “crypto winter”, with the price of Bitcoin under 20,000 US dollars and only a few days after the dramatic events of leading “De-Fi” (decentralised finance) projects Luna and Celsius collapsing, causing a domino effect in the broader “crypto economy” (Oliver 2022), NFTs are still turning over millions daily. As of the 19th of June, 2022, the 24-hour NFT trading volume is more than 37 million US dollars (Anon. 2022c). The financial resilience of NFTs, despite the dramatic events unfolding in the broader space, could be attributed to their core value, i.e. the scarcity value. Similar to how real-world collectors invest in paintings, rare watches, trading cards or any other type of collectibles, digital natives invest in NFTs. Nevertheless, a decline in the NFT market is expected. According to Gartner Research, in August 2021 NFTs were at the “Peak of Inflated Expectations” right before the "Trough of Disillusionment”; the firm projected that it will take 2 to 5 years before the technology reaches the “Plateau of Productivity” (Gartner 2021).

Resilience in the bear market and millions in daily trading volume for such a nascent technology, highlight the potential of NFTs. However, despite the galloping pace in which the digital sector is evolving, NFTs remain a raw medium, plagued by risks that must be acknowledged, navigated and addressed with caution. The main criticism against NFTs relates to their environmental impact. In today’s format NFTs were invented on the Ethereum blockchain network, which is also where the vast majority of NFTs are still being traded (Coin Telegraph Research 2021). The energy consumption of Ethereum is indeed substantial. Beeple’s auction by Christie’s alone, consumed energy enough to power 15 US homes in a year (Coin Telegraph Research 2021). However, NFTs are not inherently energy-intensive. As a technology NFTs are blockchain-agnostic, therefore if they are minted and transacted on energy-efficient blockchains, i.e. Layer-1, Proof-of-Stake blockchains (such as Cardano and Algorand) their environmental footprint can be considered negligible compared to the amounts of money they help raise. The copyright landscape is also unclear, creating a confusion to creators of NFTs and collectors alike. Research indicates that, legally, NFTs are similar to paintings, where the collector only gains ownership over the actual asset (e.g. the frame and canvas in the case of paintings), but no copyright, or commercial rights over the artwork depicted (Evans 2019). However, popular projects such as the Bored Apes which are one of the best-selling “PFP” (Picture for Profile) NFT collections of all time with millions of US dollars in weekly trading volume, award full copyright to token holders (Dale, 2021), creating the false impression amongst collectors that this may apply to all NFTs. Beyond the environmental impact of the NFT trade and the lack of clarity regarding copyright, there are also some fundamental technological shortcomings of blockchain technology, when compared to web 2.0 and to what Internet users of today have come to expect from software applications. As explained in (Valeonti 2022) in the decentralised web “Undo” and “Reset your Password” are no longer an option and the absence of such safety nets is critical. Indicatively, the ZMK Centre for Art and Media in Karlsruhe, Germany, permanently lost access to 2 rare NFTs of CryptoPunks accidentally, which were valued more than 100,000 US dollars each (Batycka 2022). Regarding the inability to retrieve lost accounts, in August 2021, prominent NFT journalist Farokh shared on Twitter the link to his lost wallet, expressing his despair over being permanently locked out of it; the NFTs it contained were worth more than half a million US dollars at the time (Twitter 2022). Some of those challenges can be addressed by using custodian, centralised platforms (such as the Binance NFT Marketplace), but these also come with risks, especially during bear markets.

In the context of Policy and our cultural heritage, one of the key risks that must be acknowledged when employing NFTs is that of “digital deaccessions” (Bailey 2021). The vast majority of NFTs created and sold by museums to date, have featured images of artefacts sourced from the respective institutions’ digitised collections. However, this a practice that should preferably be avoided, especially in the case of publicly funded museums. Τhe digitised images of museum artefacts are their digital surrogates, also referred to as “digital twins” (Niccolucci et al. 2022). Considering that NFTs (being the digital equivalent of Limited Editions, as visualised in Figure 1) present a form of ownership over the asset they feature, selling “digital twins” raises issues of digital deaccessioning. National museums are not the owners of their collections, but instead their custodians, supported by taxpayers’ money. As such, it is advisable that cultural heritage institutions, and especially national, publicly funded museums, galleries, archives and historical sites should refrain from selling NFTs with images from their digitised collections, or NFTs of any other form of “digital twins” (where a 1-to-1 correlation can be made between the digital asset minted and the actual artefact). Instead, museums are advised to mint NFTs of derivative works, new content and new concepts. The NFT minted by the Whitworth Gallery that features a new image of the artwork, which was created especially for this NFT, instead of the artwork’s digital surrogate, which is part of the museum’s digitised collection, is one such example. As institutions gradually gain a better understanding of this new medium, it is expected to start seeing increasingly creative uses of NFTs, that leverage their multifaceted nature and capacity for limitless customisation, transparency, immutability and even co-ownership with F-NFTs, “Fractionalised NFTs” (Valeonti et al. 2021), helping cultural heritage organisations engage with a young, dynamic, tech-savvy audience they could not reach otherwise. In addition, to the aforementioned challenges, similar to any other rapidly expanding market, the NFT space has also been subject to fraud; there have been cases where individuals with no connection to cultural heritage institutions have taken images of artworks and sold them as NFTs, implying partnership and endorsement (Cascone 2021). Such attempts failed to raise any funds, however, as these cases are becoming increasingly common, institutions must have clear guidance on how to respond to such scenarios. In such instances, institutions must distance themselves from the NFT creators (unless they have been actively involved in the project and were informed well in advance, allowing them enough time to perform their own risk assessment and seek advice from domain experts) and follow the take-down procedure of the respective platform where the NFTs were minted; take-down policies are provided by all leading NFT marketplaces. Finally, from a collector’s standpoint, a risk to be considered is that NFTs present a speculative investment and it is uncertain how they will perform over the longer term, similar to any other investment.

Although the challenges associated with NFTs are numerous, if they are acknowledged and addressed proactively, they have the potential to make a significant contribution towards the financial sustainability of creatives and cultural institutions, helping them in the long-term becoming less reliant on public funding. NFTs and the broader domain of web3 open up a wealth of opportunity for artists and cultural heritage institutions alike, paving the way for their financial sustainability and independence, effectively saving taxpayers’ money. By investing in research, development, support and training, whilst leveraging early successes of Scottish artists, Scotland can help shape that future of the increasingly digital cultural sector.

In Scotland there have been a variety of artists who have successfully minted and sold NFTs (McMahon 2021), whilst with regards to the cultural heritage sector, so far there has been no known selling of NFTs from a gallery, library, archive, or museum in Scotland. The speed and pace of institutional decision-making may be a factor, as may the inherently risk-averse nature of museums in general, often being amongst the late adopters of bleeding-edge technologies (Valeonti et al. 2021).

Edinburgh-based artist and graduate of the Edinburgh College of Art Trevor Jones (Klein 2021) is amongst the top-selling NFT artists of all time with sales in excess of 20 million US dollars (Anon. 2022a). Jones has focussed on cryptocurrency and the crypto-art space since 2017. His first NFT creation, the “EthGirl” sold for a final bid of 10,027 US dollars, whilst his oil painting “Picasso’s Bull” was digitally adapted as an animated NFT, selling for 55,555 US dollars on marketplace Nifty Gateway. Jones has held the first place for the highest bids on leading NFT marketplaces, including SuperRare, Nifty Gateway and MakersPlace. In February 2021, Jones’ “Bitcoin Angel” set a record for the “most expensive open-edition NFT artwork” at the time, when it sold for 3.2 million US dollars (Klein 2021). “The acceleration of the market” according to the artist is both “good and bad” because "there’s more money and attention available for emerging artists, who can make more from their work because NFT platforms take smaller cuts than traditional galleries and offer royalties for re-sales”; on the other hand, the exponential growth of the NFT market according to Jones “makes it harder for those emerging or lesser-known artists to get seen” (Klein 2021).

Edinburgh-based artist Anna Louise Simpson has sold NFTs for 28,000 (McMahon 2021). A divorced, mother of two, who discovered NFTs at 47 (Ables 2021), Simpson explains the significant impact this new medium had in her career and personal life: “As a woman and a mother, the financial freedom that this offers as an artist to not be struggling to make ends meet is incredible and I think a lot of that’s been lost in some of the conversations about NFTs”; Simpson adds that “the financial impact for women and mothers is absolutely huge” (McMahon 2021). Simpson also highlights the potential positive impact NFTs could have on Scottish art, explaining: “If you’ve not been to the right art college or met the right people, or even exhibited in the right gallery, a lot of doors are not open to you. At least this way, you’ve got access to a global, international audience and that aspect of internationalism I think is really important for Scottish art” (McMahon 2021).

In the UK, the first museum-accredited NFT was minted by the Whitworth Gallery in July 2021, as an experiment examining alternative models of financing and social art practice (Harris 2021). Titled “The Ancient of Days” after William Blake’s homonymous artwork, the NFT featured a special picture of the piece, which was produced with multispectral imaging, instead of the work’s actual digital twin. Minted on an energy-efficient (i.e. Layer 1, Proof-of-Stake) blockchain (i.e. on the Tezos blockchain) the NFT was made available in an edition of 50 each one costing approximately £2,000 at the time (999 tezos), with two copies retained by the Whitworth Gallery. The volume did not sell out and exact numbers were not disclosed by the museum. However, for the gallery it was primarily a learning experience. A "At a time of great social and economic instability, the gallery [entered] into the emerging and chaotic world of NFTs to test alternative models of financing social practice" according to the institution (Harris 2021).

A more controversial case of museum-accredited NFTs, was that of the British Museum. In October 2021, the British Museum released NFTs of 200 Katsushika Hokusai artworks in editions of varying volume that featured the exact images from the museum’s digitised collections. The price points ranged from approximately £400 for NFTs of larger volumes (referred to as “common” NFTs), up to several thousands of pounds for “rare” and “ultra-rare” NFTs, whose volume was limited (Adam 2021). The British Museum did not disclose numbers of sales, however in January of 2022 the institution announced it will be launching more NFTs in similar formats; this time of 20 paintings by J. W. Turner (Harris 2022).

Three major issues arise in British Museum’s application of NFTs. Firstly, all tokens were minted and are now being traded on the energy-intensive Ethereum blockchain network, raising justifiable criticism. Since the British Museum began selling digital versions of works from its collection in September 2021, it has consumed enough energy “to power an average US home for at least 57 years” according to the Art Newspaper (Grosvenor 2022). Given that there are several energy-efficient blockchains (which require orders of magnitude less electricity than Ethereum) that support NFTs, similar to the Whitworth Gallery, the British Museum could have avoided to “[send] its carbon footprint soaring” (Grosvenor 2022). Secondly, British Museum’s implementation of NFTs uses images from the institutions digitised collections, causing concern in relation to digital deaccessioning. As explained earlier, NFTs present a form of ownership over the asset they feature and given that the British Museum is a national, publicly funded institution, it may not be entitled to sell any form of ownership over digital twins of artefacts in its collections, no matter how small, as the museum is a custodian (rather than the actual owner) of its collections. Thirdly, at a time where the colonial nature of world-history museums is under scrutiny, the fact that the British Museum chose a (commonly available) Japanese artist and imagery to release as NFTs raised issues of internationalisation and nationalism. Many wondered: Should the British Museum be promoting NFTs based on its British cultural collections, first? Nevertheless, the British Museum is undoubtedly amongst the early pioneers, who braved into this “chaotic world of NFTs” as the Whitworth Gallery described it, conducting the largest museum NFT project to date, providing us with an informative case study that allowed us researchers and fellow practitioners to gain a better understanding, advancing the knowledge of this new medium.

“How museums choose to use NFTs could significantly impact their digital transformation and have unforeseen implications that could potentially haunt them long into the future. I would encourage museums to follow Walt Disney’s advice and ‘not sign away anything they don’t know about.’” - Jason Bailey (Bailey 2021)

“The Whitworth decided to embark on this project because it wanted to think about how it could redistribute the wealth of its collections in the most democratic way. This technology offers the opportunity to open up the collections to the broadest possible audience.”

- The Whitworth Art Gallery (Harris 2021)

As a new, blockchain-based technology that achieved the seemingly impossible by introducing scarcity into the digital realm, NFTs have emerged as a powerful new fundraising medium that could potentially pave the way for the financial sustainability of artists and cultural heritage institutions alike, helping them become more independent and less reliant on public funding. Critically, although the NFT market is not unscathed by the dramatic events unfolding in the current "crypto winter", daily trading volume is still in the scale of tens of millions US dollars, demonstrating strong financial resilience for such a nascent medium. Irrespectively of the macro environment, market decline was to be expected, as according to Gartner Research, NFTs are currently situated at the very “Peak of Inflated Expectations”, requiring between 2 to 5 years to reach the “Plateau of Productivity” (Gartner 2021).

In Scotland, several artists have seen success in the crypto-art space and were amongst its early pioneers. This is an opportunity for the Scottish Government to support, congratulate, broadcast their success and encourage them to share their deep knowledge of this otherwise exotic space to fellow artists. Cultural heritage institutions in Scotland did not rush into NFTs and perhaps rightly so, given that NFTs as a medium are still at their infancy with expertise and even plain information about them being often scarce.

Despite their strong potential, NFTs remain a raw medium, plagued by risks that must be acknowledged, ranging from the environmental impact, depending on the blockchain they get minted to the unclear copyright landscape and the technical deficiencies of the irreversible nature of blockchain technology, i.e. that in a decentralised environment “Undo” and “Reset your Password” are no longer options; unless custodial platforms are utilised which could offer such options. However, the collapse of Celsius (Oliver 2022), which was one of the leading custodial wallets, highlighted that centralised platforms also come with risk. Finally, NFTs must be treated and communicated to collectors as a high-risk investment, whose return remains to be seen, given the nascent stage of the technology. Nevertheless, given the substantial potential of the technology, after taking into consideration the aforementioned challenges, the appropriate framework should be provided for institutions and individual creators to explore this new medium cautiously and methodically.

Recommendations

Given that NFTs provide a medium that in the long-term could help artists and cultural organisations become more financially independent and less reliant on public funding learning, training, familiarisation and cautious experimentation with NFTs should be encouraged.

Experimentation with this nascent, new medium should only be done after acknowledging and proactively addressing associated risks. The two main risks concerning cultural heritage institutions are (a) the environmental impact and (b) digital deaccessioning.

Regarding the former (environmental impact), only use of environmentally friendly NFTs must be encouraged. Environmentally friendly NFTs (also referred to as “Green NFTs” or “Clean NFTs”) should only be considered those minted on energy-efficient blockchains, which, in general terms are the Layer-1, Proof-of-Stake blockchain networks, such as Algorand and Cardano.

Regarding the latter (digital deaccessioning) cultural heritage organisations are advised against making NFTs of “digital twins” of artefacts in their collections. They are encouraged to use content, which is inspired by the artefacts instead, or content that is entirely new.

The potential of NFTs is substantial, both economically, and also in terms of outreach, enabling artists as well as institutions to engage with a global, dynamic audience that would have been challenging to reach otherwise. However, information, knowledge and expertise is scarce. Therefore, in order to maximise long-term benefits economic and otherwise, it is critical to invest early in research and development, taking series of measures that will foster innovation and knowledge building both in the industry, i.e. the cultural sector and the broader creative industries, as well as in academia.

Since Scotland is home to pioneers of web3 and to early adopters of the tight-knit and highly competitive crypto-art space (i.e. from artists, such as Trevor Jones and Anna Louise Simpson to researchers of the University of Edinburgh’s Blockchain Technology Laboratory, who are conducting pioneering research for one of the world’s largest blockchains) it is advisable to facilitate knowledge exchange; to engage with them and encourage those pioneers and early adopters to share their knowledge, invaluable insight, learnings and methodologies with other fellow artists and researchers, as well as, with creators, cultural institutions and innovation entrepreneurs, effectively creating a community of web3 creatives, researchers and practitioners. A platform that could potentially facilitate such knowledge exchange is Creative Informatics; an AHRC-funded Creative Cluster (at the University of Edinburgh, Edinburgh Napier University, Codebase and Creative Edinburgh) fostering innovation and supporting Research & Development on the overlap of technology and the creative industries in the Edinburgh city region. Beyond facilitating knowledge exchange, the Creative Informatics cluster could also contribute to community building, through the Scottish network of creative technologists it has been developing, i.e. the Creative_Tech Scotland Gathering, which is also supported by the Scottish Government Ecosystem Fund.

Case Study: Datafication of Higher Education – Joanna van der Merwe, Melissa Amorós Lark and Grégory von Boetticher

Higher education encompasses various institutions that provide tertiary education leading and are authorised to grant degrees by the Royal Charter or under Act of Parliament (Scottish Statutory Instruments, 2018). Within Scotland there are 19 higher education institutions, 16 of which are universities providing education through campus-based teaching, distance learning, educational partnerships (European Commission, 2021). These higher education institutions are grouped into four main types namely 1) Ancient Universities, 2) Chartered Universities, 3) Post 1992 Universities and 4) Small Specialist Institutions (European Commission, 2021). Scotland’s universities rank among the best in the world delivering ‘economic and social benefits for both Scotland and the wider world’ (Scottish Government, n.d.a). Investing and growing the education sector is part and parcel necessary for the Scottish Government’s progress towards the United Nations Sustainable Development Goal (SDG) 4 on quality education this education infrastructure. As well as, creating a ‘more educated, inclusive and innovative Scotland’ (Scottish Government, n.d.a; The Scottish Government, 2020).

The right to higher education is enshrined in Article 26 of the Universal Declaration of Human Rights and SDG 4 aims to ensure that this right is recognised across all groups through ensuring increasing access to education (OHCHR, n.d.; United Nations, n.d.). The Scottish Government has included this focus on equal access in its 2021 National Improvement Framework and Improvement Plan, as well as, in the Scottish National Framework (The Scottish Government, 2020; Scottish Government, n.d c). Furthermore, in relation to progress on SDG 4 the government has set the target ‘that by 2030 20% of students entering university will be from Scotland’s 20% most deprived background’ (The Scottish Government, 2020).

One of the key opportunities in achieving accessibility to higher education is the integration of digital technologies. Digital technology is being integrated into higher education at an exponential rate, especially as the COVID-19 pandemic moved teaching online causing severe educational disruption across the globe. The increased use of digital technology therefore corresponded with a rise in the amount of data collected by not only education institutions but also by private sector providers. This includes novel data (for the education sector) such as biometrics and behavioural data that can be used to gain a deeper understanding of individual students.

These digital tools and the data provide a great opportunity to improve the quality of education provided by universities and teachers. Used correctly, the tools can make education more flexible and accessible, and insights gained from data can help, for example, teachers understand whether their course material or teaching methods are effective. However, there are several risks that access to so much data presents, especially with regards to the students. Clear rules and policy guidance still need to be put in place both at the government and at the higher education institution level. Rules and polices would have to cover topics such as who owns this data, and how such ownership is practically implemented, who can use this data and for what purposes, how to achieve transparency about the data collection and involve data subjects in the dialogue on what the rules and guidance will look like.

If done correctly, these tools and data can change the way higher education is delivered, ensuring that it is flexible and accessible. It can also empower students and teachers by giving them insights into their learning and teaching practices allowing them to improve themselves and their learning/teaching journey. As well as using digital technologies to use digital technology to increase community engagement and participation. Conversely, if implemented without adequate ethical considerations this data can be used to create a model of education reliant on the constant surveillance of teachers and students rather than empowerment. Additionally, their data being used for-profit motives, their right to privacy challenged and boundaries between personal and professional/student life blurred.

Education has always relied on data. At a fundamental level, education institutions have kept track of students’ progress by grading them with a numerical value (or letters). In that sense, data in education is nothing new. However, the last 10 years have brought about an explosion in not only the (digital) tools needed to carry out education, but also the data they produce and ways it is collected. The data created is not just about the student themselves, such as their admission files and course records, but also how they behave and engage with material throughout their learning career (Atif, et al., 2013). Moreover, the increased dependence on tooling and the use of third-party companies from around the world also complicate issues like data ownership, which now depend on case-by-case contract negotiation moving away from higher education institutions’ traditional data ownership (Regan & Jesse, 2019). Renewed attention should be paid to make sure students are protected from any potential negative impact based on data collected during their educational journey (Barrett, 2021).

It is almost impossible to find a higher education institution that does not make use of, for example, Learning Management Systems (LMS), learning analytics, plagiarism detectors or automatic grading software. The reasons for the introduction of such technologies are varied as they were put in place to aid with ‘improving retention, addressing curriculum standards, increasing accountability, measuring teaching quality, graduation rates and employment placement’ (Arnold & Pistilli, 2012; Dawson, 2011; Kovačić, 2012; Atif, et al., 2013).

Many benefits are brought about by using digital tooling education; however, it is important to be cautious about touting possible advantages as we are only at the starting point of this technological development. This section will point out three main opportunities: first, administrative ease, as technology relieves administrative pressure on education stakeholders, including teachers, students, and support staff, by improving administrative services’ efficiency and reducing costs (Bichsel, 2012) Data generated from the tools also allows for better forecasting and more effective resource planning (Gašević & Dawson, 2015). Second, quality enhancement, as it allows teachers to receive feedback on assignments and courses. They can take those insights and improve their work through a process termed ‘academic analytics’. Lastly, giving students access to personalised insights from learning analytics allows them to monitor their education progress, both at the course and programme level, supporting reflection and allowing for timely interventions.

However, the risks brought about by digital technologies should not be cast aside by their apparent benefits and opportunities. Three major risk categories will be presented below. First, the use of digital technologies if not done correctly can be drivers of digital inequalities and exclusion also known as ‘digital poverty’, as they can mirror and increase pre-existing vulnerabilities present in societies and erase any accessibility the technologies claim to facilitate. Second, student vulnerability and lack of privacy as well as increased surveillance, as studying puts them under the ‘all-pervasive gaze of their institution’ (Prinsloo & Slade, 2016; Knox, 2010)that analyses and collects all sorts of data throughout their education journey. All this data collection leaves students vulnerable to social and demographic profiling with untold negative consequences if poorly managed. And finally, compounding these vulnerabilities is the fact that students have little power to consent or workaround technological developments due to the uneven power dynamics between them and the higher education institutions pushing for digital transformation. However, institutions are also vulnerable themselves when negotiating against tech giants regarding tooling, security, and data ownership.

All in all, we would be remiss to not mention the benefits that have been brought about with the introduction of digital tooling in education. However, no two technologies are the same and the use of automated decision-making tools along with surveillance technology should always be scrutinised, as the consequences of misuse are of danger not only to institutions but especially to the students for whom they have a duty of care.

With the datafication of education, reflecting on the ethical concerns raised with previous technologies is vital in assessing the risk and benefits involved in future technologies and their implementation.

The case examples will serve to illustrate the ethical issues revolving technologies, namely digital proctoring and the video-conferencing platform Zoom, which are currently in use and have permeated the (digital) education sector in the past two years. Given the datafication of education, reflecting on the ethical concerns raised with previous technologies is vital in assessing the risks and benefits involved in future technologies and the design of a national policy on digital ethics.

Digital Proctoring

Digital proctoring, also known as digital invigilation, is the act of monitoring students through their personal computer while they complete an exam/assessment. Though, it may simply be regarded as a digital extension of a monitoring students when taking an exam in a classroom, there are many more ethical concerns that arise with digital proctoring. Indeed, as the University of Edinburgh points out: ‘[digital proctoring] should only be used in specific use cases and not as a blanket solution […]’ (University of Edinburgh, n.d. a). This weariness of digital proctoring can largely be traced back to ethical concerns that are brought about by using technology as it involves room scans and AI based invigilators, among other capabilities. The introduction of digital proctoring has brought about ample controversy, especially among students, some calling to ‘stop video monitoring’, to ‘keep tabs on students’, and to ‘protect students' privacy’ (The New York Times, 2020). Such discussions, particularly the privacy concern, were exacerbated by instances of data security breaches, such as Proctorio’s, one of the leading providers, CEO posting student's chat logs on Reddit (Zhou, 2020). Such instances highlighted that with increased use of digital technologies and data ownership being split between institutions and companies, the likelihood of security breaches augmented. Thus, university administrators have had to grapple with the risk and benefits of proctored exams. Some questioning the scalability of digital proctoring, considering technical trouble shooting, which could cause immense stress to students and teaching staff (Howie, 2020). Moreover, ‘discrimination that [has] been shown to impact the most vulnerable first’ (Foulkes, 2020) is furthered by these technologies, hindering the social mobility education is supposed to enable. In one of many forms, discrimination could occur due to the well-documented racial bias of face recognition/matching software, resulting in more people of colour being wrongly flagged for cheating than others. Alternatively, being of a poorer socio-economic background is likely to cause more stress, as affordable technology is of lower quality and thus more prone to technical issues while sitting an exam.

Yet, there are instances where digitally proctored exams are the best form of assessment, ‘for example resits by students not on campus […], smaller online courses, low stakes assessment that require invigilation’ (Blaney & Howie, 2020). Digital invigilation is effectively the translation of an offline activity (exam hall invigilation) to an online activity. In a university environment, it has several practical benefits such as flexibility of location allowing students to participate in proctored assessments all over the world and reduced physical space requirements for the University.

Zoom

The topic of digital proctoring, concerning with video-monitoring, raises wider ethical considerations for video-conferencing tools, such as Zoom. Video-conferring tools, which enable video calls of larger groups, have become invaluable for online teaching, as the primary tool for classes to convene and lectures to be given online.

Though existing before the COVID-19 pandemic, Zoom rose to prominence due to the switch to remote work/teaching/studying, in the process producing controversy about encryption, data storage and access, as well as hijacking. Most blatantly, Zoom was caught ‘sharing data […] with Facebook without the user's consent’ (Duffy, 2021). Zoom accordingly removed the elements of its software that enabled the sharing of data with Facebook, and despite previous data concerns, the daily user numbers for Zoom grew exponentially. This prompted Zoom to improve its encryption protocol, though it at first refused to install end-to-end encryption for free users: ‘[…] the implication that Zoom would only protect users' conversations from law enforcement if they were paying customers […]’ (Duffy, 2021). While Zoom eventually agreed to provide end-to-end encryption to all users, albeit free users would have to verify their phone numbers, with some of the company's servers being in China as well as the USA (Zoom, 2021). Both locations are of concern as the respective governments maintain powers, which allow them to access the data if they determine the need.

Beyond data storage, who has access to Zoom meetings is a further point of contention. Though this is mostly in the hands of the users, as they can choose to set passwords and how to share the link for the Zoom meeting, reporting users for hijacking meetings and potentially getting them banned from the platform was only introduced later. But even with enabling report features, hijacking of meetings has continued, creating horrible instances of hate speech, as hijackers are emboldened by the online environment, leading to: ‘Black and LGBT Edinburgh University students [being] attacked in Zoom meeting’ (Hunte, 2021).

As a result, many Scottish universities continue to recommend using Microsoft Teams or other platforms for discussing sensitive topics, as Zooms data protection has lost trust with many users (University of Glasgow, n.d.; University of Edinburgh, n.d. b).

Though Zoom remains a widely used video-conferencing tools with a plethora of benefits, the risks outlined above have created considerable push back and even the prohibition of Zoom at some institutions. The case of Zoom, as well as digital proctoring, highlight that without discussing the ethical concerns of technologies effectively everyone involved in digital education, may it be as a student, teacher, staff etc., is being put at risk without their awareness a fully informed consent. The datafication of education presents great benefits that will transform the future of education, yet its ethical parameters must continuously be re-examined, to not put individuals at risk, as has been the case with Zoom and digital proctoring.

To explore the current state of digital technologies and data within Scottish higher education institutions, the authors spoke to several teachers and experts, as well as a student representative, based at Scottish institutions. From those conversations, they saw several central themes emerge, namely; awareness, data and digital literacy, the need to rethink education, new divides and barriers, funding, and power dynamics between institutions, staff, and students.

Awareness Traps

Dr Michelle Olmstead, Director of the Centre for Innovation at Leiden University, highlighted the general lack of attention being paid to the increasing amount and sensitivity of data being collected in higher education, noted that ‘we do not know what the data is, we do not know what is being used and what is not’ ‘(Olmstead, 2021). Mia Clarke, Vice President Education of the Glasgow University Students’ Representative Council, pointed out another one of the challenges connected to the data collection: the lack of awareness challenge among students. She highlighted that as a student, I would not have thought about all this data about me on Zoom’ (Clarke, 2021). Dr Anna Wilson, Lecturer in Lifelong Learning at Stirling University, pointed out that ‘when it comes to data privacy for students, although we are all trained in UK GDPR compliance, we are probably not even aware of any [higher education]-specific policies, and we may not be aware of the data privacy risks posed by some digital tools’ (Wilson, 2021). This lack of awareness, alongside the novelty of the technology and the pace at which the environment is changing, creates a sense of fear amongst policy makers and ‘if the policymakers are fearful and do not know where to start, they cannot actually tell the government what they need’ (Olmstead, 2021).

Data and Digital Literacy

When discussing the implementation of data protection, one cannot avoid the topic of digital literacy, and this was evident in the expert discussions. Dr Keith Smyth, Professor of Pedagogy and lead of the Learning and Teaching Academy at the University of the Highlands and Islands, believed that ‘a key consideration of moving forward is around the data literacy’ (Smyth, 2021) dimension of digital literacy. He highlighted that ‘we have got to look at embedding data literacy into the curriculum […] the same way we have been talking for 20 years about embedding employability skills’ (Smyth, 2021). Dr Wilson warned that there is a danger in mistaking literacy with being an expert ‘where it is really not fair to expect them to be experts. You do not expect somebody to be a mechanic to drive a car, but you do make sure you design a car in a way that somebody who is not a mechanic can drive it’ (Wilson, 2021). This user-centric design approach should also apply to the use of technology in higher education, particularly to understand the data being collected and used. Second, when it comes to data literacy, those interpreting the data and basing policy decisions on it, or teachers taking actions on it, must understand that ‘a quantitative figure does not really tell you everything’, and that data-driven decisions must come in conjunction with analytics. As Mia Clarke concurred, ‘yes, analytics [are] important, but a part of the whole’ (Clarke, 2021). Moreover, relying on analytics alone is extremely dangerous – especially if policies leave little to ‘no consideration for the human condition’ (Clarke, 2021). In his work Dr Smyth and his colleagues are actively working against that; he explained that they are taking a holistic approach to analytics, where analytics themselves are not considered equivalent to student engagement or learning in themselves, but a part of the bigger picture.

Rethinking Education

The experts highlighted that a major challenge around the use of digital tooling and data in education is to ensure that their use is based on a strong model of education. Building on a strong model of education is an ethical requirement; however, it may also indirectly address ethical challenges such as those that exist around online proctoring. Dr Ben Williamson, Chancellor’s Fellow at the Centre for Research in Digital Education at the University of Edinburgh, pointed out that ‘if education has moved increasingly online and is hosted on platforms that are built on fairly weak models of education then we might not be providing the best education to our students’ (Williamson, 2021). In relation to the use of Learning Analytics, Dr Wilson went further, to say ‘unless we go back to a behaviourist model of learning, then counting interactions within an online module is never going to be a good indicator of learning. We understand learning as so much more complex both socially and psychologically, and none of that gets measured by click rates and wait times and so on. [...] These quantitative behavioural measures are never going to give us information about really deep, conceptual learning’ (Wilson, 2021). Dr Smyth makes a key point in that any policy about how to ethically implement digital technology for education must be driven by practice, and so governments must be willing to ‘look to [the higher education] sector for direction because that is where practice [is] driving policy’ (Smyth, 2021), especially in this new area of data-driven education. Furthermore, Wilson, Williamson and Smyth agreed that curriculum updates are needed to accommodate these challenges and opportunities. In line with the curriculum updates, Wilson also highlighted the opportunities that digital education gives rise to, particularly shifting to skill rather than degree focused university programmes and increased accessibility no matter the socio-economic background or geographic location of an individual. '[...] Build up a portfolio of qualifications and that is something that can be enabled through digital tools and digital learning [...] in the way that courses are provided, allowing more remote access, and allowing more remote learning, but also continuing with face to face and on-campus learning.' (Wilson, 2021)

New Divides and Barriers

Although digital technology is seen as an opportunity to increase accessibility and reduce divides, the experts pointed out that it is creating new divides, and old divides simply manifest in new ways. Mia Clarke noted that minorities and those already disadvantaged are marginalised even further: one thing that you can notice when students enter a call with their camera on is the quality, which is normally accredited to how expensive the laptop or computer is. So, I think it definitely offsets people already when they turn cameras on and speak as you can already tell if they are from a marginalised group. I think digital education exacerbates that divide more than in person teaching’. There is a lack of recognition that the divide is now based on who has access to what quality of technology, something that is evident when interacting in the online environment. Dr Wilson actively tries to keep the digital tooling in her education as simple as possible based on the recognition that some of her students are on the poorer end of the digital spectrum. Dr Smyth argued that the COVID-19 pandemic revealed some key elements of how this divide manifests when students ‘chose external technologies to try and self-organise […] the students that did not know [about the technology] or who are not on Facebook or WhatsApp got left behind’.

Furthermore, Clarke highlighted a specific challenge for gender diverse students and the collection and storage of their data: ‘I think there is a huge issue in terms of recording and video software, especially for gender diverse students who [seem to be] transitioning. They might struggle to present on camera or speak knowing someone has a recording of their voice pre-transition. That can be really damaging to them’. Keith Smyth remarked that at the University of the Highlands and Islands, the work with students has gone ‘beyond simple representation on things like the steering groups’ (Smyth, 2021) but rather engaged them in the process of identifying the ‘initial set of student engagement indicators that they were happy with […] and which they felt would support their learning and teaching experience’ (Smyth, 2021).

Funding

Experts brought up funding as one of the barriers to the ethical implementation of digital tooling in higher education, especially when ensuring resources are available for exploring the ethical challenges and requirements, as well as ensuring equality of opportunities amongst students. Dr Wilson highlighted that ‘you need effort to understand the technologies and what they do. Each new technology, and each new supplier, probably has new issues that need to be thought about’ (Wilson, 2021). Furthermore, not only does generating the knowledge and ethical guidelines take resources, but Dr Olmstead pointed out that ‘it is fairly easy to get the policy and people to agree on an ethical framework or a data privacy framework. The difficult part is actually implementing that with the funding that exists’ (Olmstead, 2021). Part of the implementation challenge that is often overlooked is providing access to the equipment needed to take part in digital education, which relates closely to new barriers and divides or biases that may result from unequal access to equipment. Mia Clarke added that ‘education has been underfunded in Scotland and […] that is a disaster because there is so much untapped potential. If you want […] students coming in from disadvantaged backgrounds, it’s great if there are going to be provisions for them’ (Clarke, 2021), provisions that are not currently available. As such, funding is one of the key steppingstones that allows for as Richard Lochhead, the Higher Education Minister, said for ‘every young person in Scotland [to have] an equal chance of success, no matter their background or circumstance’ (Denholm, 2019).

Power Dynamics Between Institutions, Staff, and Students

Higher education institutions inherently have a baseline of power based on their position as the issuer of the degree and provider of the education. However, digital tooling and the types of data that are now available hold the potential to create a dynamic where students are especially at risk for losing their power to protect what data is collected about them, how it is processed, for what purpose it is used and what conclusion/assumption is made about them based on that data. Signs of this unequal power dynamic are clear when discussing the current levels of informed consent at institutions. Dr Wilson commented that universities ‘do informed consent incredibly badly because basically you do not have access to the degree if you do not agree to the terms and conditions [which] also discourages people from finding out what they are’. Dr Williamson noted that ‘every institution is running a learning management system (LMS), and really by signing an agreement that you’re going to attend the university […] you are opting into the [LMS]. It’s more or less impossible to attend university without it. There is no meaningful consent, you are just in’ (Williamson, 2021). Taking this trend further, he indicates that: ‘if you choose to go to a university and it [uses] Amazon Web Services (AWS), and that means that every data point that is generated in that institution goes to a data lake, which is hosted by AWS, then your choice to go to that university means you are consenting for your information to go to Amazon’. This demonstrates how this data then brings Big Tech into education data. There are ‘cautionary tales of taking a market focussed approach which assumes that the business deal or financial agreement is always in the best interests of the institution’ (Williamson, 2021)

Without informed consent and adequate student involvement, there is also the concern that universities will use data about students combined with unfounded assumptions, leading to incorrect conclusions. Mia Clarke provided a pertinent example of what this may look like, based on a pilot that was conducted: ‘you can look at analytics in terms of how often they engage with online content, but then add things into it like postcode or the Scottish Index of Multiple Deprivation (SIMD). To include things like [postcode or SIMD] puts an expectation on certain students that might not even exist. Just because someone comes from that SIMD background, that doesn’t necessarily mean that they are at high risk of dropping out’ (Clarke, 2021).

The barriers to ethically implementing digital tooling in higher education are plenty; however, there are also many opportunities and benefits. As Mia Clarke put it, ‘digital education is the future of education in Scotland, and for the wide world’ (Clarke, 2021). She cannot see things going back to the way they were and hopes that the opportunities such as accessibility and openness are tapped into further.

Summary, Mapping, Recommendations

These recommendations focus on broader-ranging steps that must be taken to ensure that as the datafication of education continues, it follows a route that enables the establishment of strong ethical foundations. The COVID-19 pandemic provided key insights into the vital role digital technologies play in ensuring access to education, amongst other opportunities highlighted throughout the case. It also clearly demonstrated that ethical challenges around these new technologies, the types and amount of data they collect, the automated decisions based on the data (e.g., exam proctoring), risks around use of tools without adequate protections (e.g., zoom) and the inability to access education without these tools (making it harder for students to opt out). These recommendations are broader than specific tools as the education sector [is at a nexus] where it can build the ethical foundations for a future of digital education. It is key that the Scottish Government take this opportunity to set up these foundations, and do so as quickly as possible, will reduce the monumental task of making technologies and institutions more ethical later down the line. Based on the research and insight gained throughout the process the authors have come up with this non-exhaustive list of recommendations:

  • Including higher education, experts and stakeholders have a meaningful contribution to the identification and development of necessary policy.

Perhaps the most important aspect of the future development of digital tooling and data-driven processes in higher education is ensuring the policy decisions are driven by the right people. The policy must be sector driven, as the complexities and expertise are within the institutions. Furthermore, the current level of student involvement at all levels of policy development is insufficient with students consistently being “outnumbered” in representation but also, the few students who are involved are expected to provide input across multiple topics at various levels. Individual universities, such as the University of the Highlands and Islands, have made significant steps in ensuring that a sufficient portion of the student body are involved in decision-making around data reliant developments such as learning analytics. The processes that these universities use need to be obligatory across the sector, linked to QAA Scotland and, aligned with Scotland’s Quality Enhancement Framework (QEF).

  • Programs and funding support are needed.

Support, in the form of funding and/or special programs, must be established, or expanded if already existing in inadequate amounts or size, to ensure adequate access to tech infrastructure and equipment to effectively participate in increasingly digitised education. This is indirectly and directly linked to ability to protect data but also ensure fair representation should systems begin to base decisions on this data.

  • Establishment of sector wide standards and codes of conduct, especially ethical codes.

Shared standards for the ethical use of data-driven technology and decision-making must be established for the higher education sector as a whole. This includes, ethical codes of conduct, methodologies for evaluating tools, technical standards, data, and digital capabilities (within institutions) etc. This can also include a suite of tools that are contracted at the government level and made available as standard tools to all institutions to ensure that the standards are met but also balance out power dynamics between private sector providers and the institutions. This will facilitate integration of tool across the sector, a key aspect of enabling flexibility around data collection (especially variations between what students opt-in and opt-out of).

  • Updating of core competencies that all students must develop.

The core competencies of all curricula, that must be included across all education, need to be broadened to include data and digital literacy. The definition of what it means to be literate in these areas also needs to be clearly delineated. Being data and digitally literate does not mean that students must develop deep technical expertise (e.g., programming skills) but rather understand it to the extent that consent can be truly informed, they meaningfully participate in dialogue and understand the wider implications of the data generated and used. During the current transition into increasingly digital education, the baseline levels of knowledge will vary so the manner in which this is delivered will need to evolve over time (Wilson, 2021; Smyth, 2021).

  • Identification of new compulsory position, responsibilities, and processes.

As more data-driven technologies and processes are integrated into higher education (and the scope of their impact widens), the institutions take on increasing responsibility for ensuring that it is done so ethically. However, a challenge within these institutions is that the capacity to evaluate the ethical risks and oversee the ethical implementation is lacking. Like the creation of the Data Privacy Officer position that was created under the UK GDPR and Data Protection Act 2018, a similar position for ethical implementation of data-driven technology and decision-making within higher education institutions. This position must be connected to an ethical review board that takes the evaluation of the tools beyond a data privacy assessment, but includes assessing impacts on accessibility, fairness, discrimination etc. This mechanism and process must be made compulsory protect students and their data and to ensure all tools rolled out are done so based on well-grounded and researched evidence of the effectivity.

We would like to thank the contributing experts for sharing their valuable time and opinions with us which have been key to shaping this case study. We would also like to thank the organising team for their support as well as the trust they put on us.

Experts consulted for this case study:

Dr Anna Wilson – Lecturer in Lifelong Learning at Stirling University

Dr Ben Williamson – Chancellor’s Fellow at the Centre for Research in Digital Education, University of Edinburgh

Mia Clarke – Vice President Education at Glasgow University Students’ Representative Council

Dr Keith Smyth – Professor of Pedagogy and Head of the Learning and Teaching Academy

Dr Michelle Olmstead – Director, Centre for Innovation Leiden University

Case Study: Fintech in Scotland – Felix Honecker

Over the past decade, innovations in financial technology (fintech) have started to transform the way financial services are delivered. Advancements in cutting-edge technologies such as artificial intelligence, natural language processing, cloud computing, Application Programming Interfaces (API), and blockchain continue to change how businesses in the sector operate, collaborate, and transact with their customers, regulators, and other stakeholders. Additionally, Open Banking practices allow fintech companies to access vast amounts of previously unavailable data (including transaction data), enabling them to develop new products and services that are potentially better suited to the needs of consumers.

Fintech’s economic potential is staggering and investment into the sector is booming, with venture capital funding in 2020 exceeding $4.1bn in the UK alone.1 Globally, the sector is projected to reach a value of about $305bn in 2025 – growth that is fuelled mainly by consumers and small and medium-sized businesses turning to fintech for payments, financial management, and financing.2 When promoting the fintech sector, businesses, NGOs and policy makers alike have highlighted not only its huge economic potential, but also its capacity to trigger positive social change. Fintech can improve the efficiency and reduce costs of the current financial system, extending financial services to previously unserved or underserved households. However, consumer experts have identified several obstacles that could reduce the positive social effects of fintech and, potentially, leave consumers worse off.

Convenience, speed, lower cost, and simplicity may come at the expense of losing control over our money and data, a reduction in privacy and exposure to unfamiliar security risks. And while increasing digitisation theoretically widens access to financial services by tearing down geographical barriers, it neglects that particularly vulnerable consumers often lack the capability and access to mobile or broadband data that is necessary to make use of most fintech products. These issues create ethical tensions that require thorough consideration to prevent fintech from exacerbating rather than alleviating financial and social exclusion. The UK have established themselves as global leaders in fintech and aim to maintain that position by adopting specific policies and regulatory approaches that create an enhanced and enabling environment for fintech businesses. As a major financial technology hub and home to more than 140 fintech firms, it is crucial for Scotland to develop an in-depth understanding of the ethical challenges associated with facilitating the integration of technologies into financial services.

There is no doubt that fintech can play a key role in delivering some of the National Outcomes set by Scotland’s National Performance Framework. Scotland is home to a fast-growing fintech sector that benefits from the country’s historically strong financial services expertise, world class universities and talent, and excellent business support ecosystem.3 The Scottish cluster has positioned itself among the leading fintech destinations in the UK and the world, creating high-income, tech-based employment in the country’s fintech cities Edinburgh, Glasgow, Aberdeen, Dundee, Stirling, and Perth.4 Through an increase in remote work, current and future jobs in fintech are no longer restricted to these larger cities but open up opportunities for high-quality employment across Scotland. Moreover, companies across a varied range of sectors can adopt financial technology tools for payments, accounting, cash flow management, smart contracts, and other business functions to increase productivity. These developments, therefore, contribute to the Economy, International, and Fair Work and Business outcomes of the performance framework.5

Additionally, fintech has the potential to significantly improve social inclusion and tackle poverty. Digital finance is faster, more efficient, and usually cheaper than traditional financial services. Building on these advantages, fintech firms are extending access to financial services to low-income households and small businesses that previously had been unserved or underserved.6 World Bank data shows that between 2011 and 2017 about 1.2 billion people have gained access to a transaction account, with much of this progress attributable directly to digital technologies.7 In addition, open banking practices combined with AI and machine learning have enabled fintech firms to use new, supposedly fairer approaches to credit scoring and risk assessment that are more transparent and do not solely rely on a credit history.8 These technologies help financial businesses to gain an improved understanding of their customers and allow them to provide better money advice or develop products that are more suitable to their customers’ (or excluded households’) needs. InBest.ai, for example, use artificial intelligence to transform a variety of banking data into a holistic picture of each customer. Financial institutions use InBest’s platform to uncover early warning signs of financial vulnerability based on personal circumstances, financial situation, and transaction patterns. Financial institutions use these insights to provide customised solutions that improve the financial wellbeing and resilience of consumers.

By facilitating access to accounts and credit, fintech is creating opportunities for wider sections of society to participate in formal economic activity. Moreover, complementary products and services, such as benefits calculators or AI-enabled financial management apps, can help consumers reduce expenditure, maximise income, and manage their money more effectively. These tools can play a decisive role in debt and financial poverty prevention. Reduced complexity and methods like gamification (i.e., the use of game concepts and design principles for non-gaming applications) can further improve how people interact with their financial futures and increase financial literacy. Scottish fintech firm Sonik Pocket, for example, uses gamification to teach children the value of money and help them achieve savings goals.

Fintech, therefore, has the potential to create income and employment, enhance economic growth, narrow income inequalities, improve financial literacy, and significantly reduce poverty and social exclusion.9 And there is still plenty of scope to improve the positive social impact of fintech, particularly through expanding e-government services and targeted fiscal measures. However, there are also risks related to higher fintech adoption. Careful ethical considerations and comprehensive policy interventions are required to realise fintech’s economic and social potential while also mitigating these underlying risks.

Notwithstanding the aforementioned opportunities, there are several challenges emerging from increased fintech adoption. Social inclusion could be at risk if the ongoing COVID-19 pandemic accelerates the transition to digital financial services. Unequal access to digital infrastructure (e.g., lack of access to mobile phones, computers, or the internet) could exacerbate existing and create new forms of exclusion. Additionally, people who may have access to the necessary infrastructure but lack digital expertise may refrain from using new technologies and miss the opportunities associated with them. This could aggravate existing inequalities, expand the digital divide, and further isolate vulnerable groups.10

Similarly, machine learning and data biases (which occur due to prejudiced assumptions made during the algorithm development process or biased training data) as well as inaccurate or incomplete data could restrict rather than broaden access to credit. There is some evidence indicating that, instead of delivering on the promise of facilitating access to affordable credit for previously excluded consumers, increasing data points in credit scoring also increases inaccuracies which in turn negatively affect creditworthiness.11 People who deliberately avoid leaving a data trail or who suffer from data poverty12 will be disadvantaged by these approaches if there are no moderating measures in place.13 This potentially forces people to establish a data history at the expense of their privacy if they want to avoid being subjected to unfair price discrimination (or being excluded from credit altogether). At worst, some of our most sensitive and private data could be extracted and used for exploitative ends. Thereby, the financial inclusion argument could be co-opted by firms to bolster their legitimacy.

Comparable risks emerge when fintech applications are “gamified.” While gamification can positively shape and encourage better financial behaviour or improve financial literacy, it can just as easily entice people to spend over their budget or aim for short-term rewards rather than make the decisions that are in their best interest in the long term. Some investment and trading apps, for example, have drawn huge criticism for using gamification to encourage their users to trade more frequently even though this approach often increases the likelihood of losses.14 Young, inexperienced consumers with low financial literacy tend to be particularly susceptible to such unethical applications of gamification. Similar challenges exist with businesses offering easy access to micro credit (e.g., Klarna’s buy-now-pay-later or instalment options), which often entice young people to spend over their budgets.15

Recommendations

The government must ensure digital financial inclusion, for example by providing means and funding for digital infrastructure (devices, internet access, and special equipment for people with disabilities), provide skills training, and build trust and motivation to use financial technology. Since there will always be varying degrees of tech-savviness and differences in the willingness to engage with technology, the government should also support viable alternatives to high-tech digital finance. Promoting and financially supporting the digitisation of credit unions’ and building societies’ operations, for example, could allow people who are not able or willing to engage with fintech solutions to still benefit from its advantages (lower cost, speed, more transparent processes etc) without personally adopting new tools.

To maximise the positive social impact, the government should facilitate and support long-term dialogue and relationships between financial technology firms and organisations that have experience working with and are trusted by vulnerable consumers (e.g., NGOs, credit unions). This will facilitate innovation and improve fintech tools for financial inclusion through expert input while also accelerating adoption of such tools if trusted institutions recommend them to the people they work with.

The government and financial regulators should require transparent and understandable explanations on what the collected information is used for, why it is required, with whom it is shared for what purposes, and how it is protected to ensure data privacy. There could, for example, be terms and conditions templates that follow the same format, use similar language, have standardised components etc, to improve understandability and avoid illegitimate or unethical data practices being buried in complexity.

To complement (3) and to enable sound decision-making when it comes to data sharing, the government should facilitate a balanced education process for citizens to better understand their digital footprint, put them in control of their own data, and recognise both the upsides (e.g., source of useful innovation, products and services tailored to specific needs) and downsides (e.g., loss of privacy, potentially loss of control, increased susceptibility to manipulation of behaviour) of sharing their data. A better understanding will allow consumers to make informed, case-by-case decisions on what data to share, with whom, under which conditions.

The government and regulators could require technology firms to set the strictest privacy settings as the default, so that consumers must give explicit consent to all aspects of data collection. Accepting/rejecting should be possible through a transparent and understandable process without “convenience traps” (no “accept all” option to quickly get through the process). This would create an incentive for businesses to adopt data practices that are acceptable to consumers, and to clearly demonstrate to consumers how sharing their data will create value for them or others.

Contact

Email: digitalethics@gov.scot

Back to top