The Digital Ethics Officer

A new role to ensure digital ethics within an organisation.

The need of ethics for AI

The success of initiatives using artificial intelligence such as ChatGPT, Bard?or Lensa, recently very popular,?moves towards the democratisation of AI, allowing a more natural and human interaction than all the existing search engines on the market.?

However it also makes everybody be careful because they deliver a seductive, engaging, natural speech ( in the case of #ChatGPT or #Bard), but where is not easy to perceive if what is said is trustworthy.?

It is possibile therefore to create?also big disappointments because it can say things that are deeply false.

If we have no critical judgment, we are heading for problems.?

For years, we have had an approach that says that AI is there to assist and enhance the human. Not replace it. ChatGPT/Bard reinforces the importance of maintaining control, thinking skills and free will.?

The more AI becomes powerful, the more it is necessary to know how to manage this technology.

For this, trust is essential.?

In 2022, a study of 5,000 executives around the world, made by a multinational company, showed that the notions of fairness, security, transparency and ethics were predominant. Strategic, critical subjects, shared by all. Without trust, there is no adoption, no ROI, no scaling. It’s a proof of maturity: people are asking for that trust.?

And the AI revolution will not happen without it. Therefore AI must be transparent, explainable, robust, respectful of privacy. Finally, there is a more subjective notion: ethics.?

We shouldn’t forget that we are all biased.?

We all have different education, values, beliefs, and therefore ethics. We are in a skewed world; it is therefore necessary to work to reduce the bias as little as possible. It's not AI, it's us. An AI is not omnipotent: it is as good or bad as the person who is going to use it or who entered the data. It requires a lot of effort, anyway we need to keep control in the whole creation and deployment process. The human being, or better the whole Globe ecosystem, must remain at the center of everything.

Regarding ethics, the #EuropeanAIAct, recently proposed,?will objectify things, allow the developers to be confident in the manufacturing and use processes. Behind there is already the European data protection policy that is stricter than in the USA or China, it could be perceived as a limit to development, however regulating the AI is essential.?

Some large groups have already established internal rules for the development of AI programs in safety but we cannot rely on these initiatives alone to guarantee the development of good AI. Otherwise, employees will find themselves without a frame of reference, therefore without knowing if what they are doing is aligned with the ethics accepted in the society.?

Going further many professions have a code of ethics. It is this equivalence that must be created, known to all.

We could foreseen that this European regulatory framework becomes the global framework. If we succeed in getting this accepted, it will be a real mark of leadership; and Europe can do it!?It is matter to set the means. It is a societal subject, which affects us all. Without a framework, we risk to arrive at errors and unacceptable behaviours.

It is vital because AI will be everywhere, affecting all industries, all people, all fields. From coffee machine repair to medicine. It is a real deep revolution, and as all revolution will imply change of powers and probably will there be job losses related to the development of AI.?Like any new technology, it will disrupt the job market. I will mainly talk about transformation, induced by this integration of AI. And, for that, we come back to the fundamental notion of trust. People need to know when AI was used, how, by whom? That's the issue. Because the rise of AI is inevitable: a wave doesn't stop, it needs to be surfed.

That links into the concept of AI quality and in particular when models are used to make predictions and decisions that impact the lives of people, it is absolutely essential to respect societal and legal expectations of #AItransparency, #AIfairness and #AIprivacy.?

Quite often, attributes like fairness and privacy are in conflict with simply optimising for model performance. Yet they are critical to adhere to in order to responsibly drive business value.

In short, #AIQuality encompasses not just model performance metrics, but a much richer set of attributes that capture how well the model will generalise, including its conceptual soundness, explainability, stability, robustness, reliability and data quality. It also includes attributes embodying societal and legal expectations of transparency, fairness and privacy, and process-level attributes supporting communication, reproducibility, and auditability. (See fig. 1)

The role of Digital Ethics Officer

All the above pose the problem for organisations that develop innovation through digital projects, which include the use of AI. They need to equip themselves with ethics appropriate to human values and the challenges of the millennium during the conceptualisation phase. The ethical discipline in this field is not always yet formalised within the policies of private or public organisations; sometimes as a deliberate choice sometimes induced by a focus on the objectives without considering the side effects of a specific application.

In the discussions that have recently animated the introduction of AI in many fields, the idea of an organisational role that has the precise responsibility of ensuring compliance with ethical aspects for digital projects is starting to gain ground. This is the #DigitalEthicsOfficer (#DEO).

It is definitely a new role that must work in a transversal way and with a systemic vision of digital projects, taking into account the specifics of the sector of activity, the territory on which the project will act and the related applicable ethical recommendations. A kind of conductor of a system that is both ethical, legal and technical at the service of a digital manager. He must organise, coordinate and animate the considerations on the sidelines of a digital project of a company or a public body, knowing the roles and responsibilities of the various components of the development team. Vital will be the ability to connect the various organisational areas, often organised in silos: administration, R&D, legal, production, communication, marketing, quality, customer service. Each must have its own field of intervention and the DEO must maintain a clear vision of the process, facilitating the participation of all in the definition of an ethical framework for the activities of the organisation.

It is important that he manages to maintain a common accord between the various elements regarding values to defend and ethical rules to put into practice. In fact, he has to enforce an overall vision between the various components. Both in the planning phase, Design Ethics, and in the execution phase, Evolution Ethics, which ensures its consistency over time. The latter is a constantly evolving part and is based on the ability to consider and capitalise on the evolution of ethical, legal and technical aspects by anticipating the risks associated with a project that exploits Artificial Intelligence.

The main missions of the DEO

The role of the DEO covers multiple aspects here so that we try to briefly list the main activities below. (Summarised in fig 2)

1- Have deep knowledge and understanding of the company's digital processes

2- Clearly identify exposures and ethical risks of projects under development considering all use cases, planning and prioritising actions to be implemented to minimise ethical and legal risks.

3- Organise and lead the operational governance for the supervision of the "human" relationship aspects of Artificial Intelligence projects).

4- Be the ethical reference for all digital projects (Ethics by Design/ by evolution) and for all certification activities and related acronyms.

5- Strengthen the transversality of the references and people involved in digital projects, the culture of cooperation and collective decisions to obtain co-planning and transparency within the organisation.

6- Sensitise and advise the different organisational entities (design, technicians, admin....) on the ethical impact of the role and the associated good practices.

7- Conceiving and distributing an ethical chart with the codes of conduct to be applied to digital activities and possibly tools available and educating staff.

8- Develop evaluation and control tools within the framework of internal (charters, codes) and external (state of the art practices) rules in relation to the organisation's area of activity.

9- Eliminate or minimise the ethical risks of digital projects, including discriminatory and exclusionary biases.

10- Integrate and improve the synergy of devices to protect the fundamental rights of consumers and citizens and, if missing, develop new ones

11- Organise internal reflections to evaluate feedbacks from customers/users or people involved in the project (internal and external)

all of the above requires that the DEO have great human, scientific and practical qualities to be able to instill the confidence necessary for the activities of guidance, cooperation initiatives, with mediator and pedagogical skills.

This also requires that the role be absolutely independent in relation to the staff and management of the organisation, to avoid exploitation and to foster trust from the various parties involved in the project. His action will have to allow the company management to become aware of the ethical effects of the project and to consequently take all the decisions in an informed way. (See fig 3)

An ethical vision from the not technical world

In the time that experiences the affirmation of digital technologies, immersed in our own life as if to merge with it - from algorithms to artificial intelligence - all the above drives to a serious reflection on the very value for humanity.?

We have already mentioned that on this path the European community has delivered an ARTIFICIAL INTELLIGENCE ACT to harmonise rules on artificial intelligence.?

In light of the speed of technological change and possible challenges, the EU is committed to strive for a balanced approach. It is in the Union interest to preserve the EU’s technological leadership and to ensure that Europeans can benefit from new technologies developed and functioning according to Union values, fundamental rights and principles. An #AIBoard, aiming to ensure them, has greater autonomy and to strengthen its role in the governance architecture for the #AIAct.

Several scientists, philosophers, theologians and bioethicists converged recently in Rome from all over the world for the international workshop “Converging on the person. Emerging technologies for the common good” (Called “Rome Treaty”).

There is a clear awareness of the criticality of our relationship with new technologies, and the warning?that the technological form of human experience is becoming more pervasive every day: in the distinctions between 'natural' and 'artificial', 'biological' and 'technological', the criteria by which to discern what is human and technique become more and more difficult. The question is cultural: ?It is necessary to decisively reaffirm the importance of the concept of personal conscience as a relational experience, which cannot disregard either corporality or culture. In other words, in the network of relationships, both subjective and community, technology cannot replace human contact, the virtual cannot replace the real and neither can social media replace the social sphere. And we are tempted to make the virtual prevail over the real: this is an ugly temptation?, the Pope said at the closure of the event.

The same topic has been recently discussed also in a dedicated session at the WAICF - World AI Cannes Festival with Federico Faggin , currently focusing on the same challenges through a dedicated?foundation in the USA. Interesting session organised by Europia , an institution which promotes an ethical and people-focused approach to the design and implementation of artificial intelligence solutions.

That of the human threatened by a technology left to dominate without control is one of the three "challenges" indicated during the Rome event. The other two are "the change in man's living conditions in the technological world" and "the concept of 'knowledge'".?

From the final press release of the conference “…the rapid development of technical means makes the interdependence between man and the 'common home' more intense and evident", both in the environment and in human life conditions, with effects and developments that are not always clear and predictable ?.?

For a "healthy" development it is necessary to take them into account.

In this rapid evolution, despite full regulations are not yet released, all organisations need to be prepared to prevent issues, be compliant with market expectations and in some way contribute to the creation of the necessary mind set to foster the technological evolution in support to humanity in a comprehensive and positive approach. The role of a Digital Ethic Officer,?inside any organisation, could be a first necessary step to assure AI governance towards a good AI.?

References

1. Adobor, H.: Exploring the role performance of corporate ethics officers. J. Bus. Ethics 69, 57–75 (2006). https://doi.org/10.1007/s10551-006-9068-7
2. CrossRef?Google Scholar??
3. Bayer, M., Hauser, I., Z?ggeler, A.: Zitieren nach ethischen Grunds?tzen? (nd). https://diglib.tugraz.at/download.php?id=4d2daecd474b4&location=browse
4. BBC: Alexa tells 10-year-old girl to touch live plug with penny (December 2021). https://www.bbc.co.uk/news/technology-59810383
5. Bennett, J.B.: Do colleges and universities need ethics officers? Acad. Leadersh.: Online J. 1(2), 4 (2003). https://scholars.fhsu.edu/alj/vol1/iss2/4
6. Braun, C.: Technologiegestützte Lehre vermitteln - TELucation an der TU Graz (2021). https://hochschulforumdigitalisierung.de/de/blog/telucation-tu-graz. Accessed 10 Feb 2022
7. Brookes, K., Davidson, P., Daly, J., Hancock, K.: Community health nursing in Australia: a critical literature review and implications for professional development. Contemp. Nurse 16, 195–207 (2004). https://doi.org/10.5172/conu.16.3.195
8. CrossRef?Google Scholar??
9. Chief Ethics and Compliance Officer (CECO) Definition Working Group: Leading Corporate Integrity: Defining the Role of the Chief Ethics and Compliance Officer. Ethics Resource Center (June 2010). https://www.ethics.org/wp-content/uploads/2010-ECI-WP-Leading-Corporate-Integrity.pdf
10. Coeckelbergh, M.: Artificial intelligence: some ethical issues and regulatory challenges. Technol. Regul. 2019, 31–34 (2019). https://doi.org/10.26116/techreg.2019.003
11. CrossRef?Google Scholar??
12. Dougherty, C.: Google photos mistakenly labels black people ‘gorillas’ (July 2015). https://bits.blogs.nytimes.com/2015/07/01/google-photos-mistakenly-labels-black-people-gorillas/
13. Ebner, M.: iMooX - a MOOC platform for all (universities). In: 2021 7th International Conference on Electrical, Electronics and Information Engineering (ICEEIE), pp. 537–540 (December 2021). https://doi.org/10.1109/ICEEIE52663.2021.9616685
14. Ebner, M., Sch?n, S., Dennerlein, S., Edelsbrunner, S., Haas, M., Nagler, W.: Digitale Transformation der Lehre an Hochschulen - ein Werkstattbericht (December 2021)
15. Google Scholar??
16. Ebner, M., Sch?n, S., Braun, C., Ebner, M., Grigoriadis, Y., Haas, M., et al.: Covid-19 epidemic as e-learning boost? Chronological development and effects at an Austrian university against the background of the concept of “e-learning readiness’’. Future Internet 12(6), 94 (2020). https://doi.org/10.3390/fi12060094
17. CrossRef?Google Scholar??
18. Edelsbrunner, S.: Entwicklungspapier E-Assessment an der TU Graz. Workingpaper, Graz University of Technology (February 2022). https://doi.org/10.3217/s2rpc-x5g66
19. Edelsbrunner, S., Steiner, K., Sch?n, S., Ebner, M., Leitner, P.: Promoting digital skills for Austrian employees through a MOOC: results and lessons learned from design and implementation. Educ. Sci. 12(2), 89 (2022). https://doi.org/10.3390/educsci12020089
20. CrossRef?Google Scholar??
21. European Commission and Directorate-General for Communications Networks, Content and Technology: Ethics guidelines for trustworthy AI. Publications Office (2019). https://doi.org/10.2759/177365
22. Gnazzo, P.J.: The chief ethics and compliance officer: a test of endurance. Bus. Soc. Rev. 116(4), 533–553 (2011). https://doi.org/10.1111/j.1467-8594.2011.00396.x
23. CrossRef?Google Scholar??
24. Gosch, N., Andrews, D., Barreiros, C., Leitner, P., Staudegger, E., Ebner, M., et al.: Learning analytics as a service for empowered learners: from data subjects to controllers. In: LAK21: 11th International Learning Analytics and Knowledge Conference, pp. 475–481. Association for Computing Machinery (2021). https://doi.org/10.1145/3448139.3448186
25. Graz University of Technology: Climate-neutral TU Graz. Roadmap. https://www.tugraz.at/en/tu-graz/university/climate-neutral-tu-graz/roadmap/#c367923. Accessed 17 Jan 2022
26. Graz University of Technology: Commission for scientific integrity and ethics. https://www.tugraz.at/en/tu-graz/organisational-structure/representative-bodies-for-members-of-tu-graz/commission-for-scientific-integrity-and-ethics/. Accessed 15 Jan 2022
27. Graz University of Technology: Electronic mailbox for anonymous tips (whistleblowing). https://www.tugraz.at/en/about-this-page/electronic-mailbox-for-anonymous-tips-whistleblowing/. Accessed 15 Jan 2022
28. Graz University of Technology: STS - science, technology and society unit. https://www.tugraz.at/arbeitsgruppen/sts/home/. Accessed 15 Jan 2022
29. Graz University of Technology: Sustainability Advisory Board. https://www.tugraz.at/en/tu-graz/organisational-structure/committees/sustainability-advisory-board/. Accessed 10 Feb 2022
30. Graz University of Technology: TUbe. https://tube.tugraz.at/. Accessed 10 Feb 2022
31. Graz University of Technology: Ethischer Kodex der TU Graz (October 2008). https://mibla-archiv.tugraz.at/08_09/Stk_1/EthikKodex1_0_1008.pdf
32. Graz University of Technology: Richtlinie zu offenen Bildungsressourcen an der Technischen Universit?t Graz (OER-Policy) (November 2020). https://www.tugraz.at/fileadmin/user_upload/tugrazExternal/02bfe6da-df31-4c20-9e9f-819251ecfd4b/2020_2021/Stk_5/RL_OER_Policy_24112020.pdf
33. Graz University of Technology: Stellen- und Funktionsbeschreibung (2020). internal document
34. Google Scholar??
35. Graz Universtity of Technology: Verhaltenskodex (Compliance Richtlinie) (June 2021). https://www.tugraz.at/fileadmin/public/Studierende_und_Bedienstete/Richtlinien_und_Verordnungen_der_TU_Graz/Verhaltenskodex_Compliance_Richtlinie_Deutsch.pdf
36. Greenberg, M.D.: Perspectives of Chief Ethics and Compliance Officers on the Detection and Prevention of Corporate Misdeeds: What the Policy Community Should Know. RAND Corporation, Santa Monica (2009)
37. Google Scholar??
38. Hoffman, W.M., Neill, J.D., Stovall, O.S.: An investigation of ethics officer independence. J. Bus. Ethics 78(1/2), 87–95 (2008). https://doi.org/10.1007/s10551-006-9312-1
39. CrossRef?Google Scholar??
40. Hogenbirk, S., van Dun, D.: Innovative ethics officers as drivers of effective ethics programs: an empirical study in the Netherlands. Bus. Ethics Eur. Rev. 30, 76–89 (2021). https://doi.org/10.1111/beer.12310
41. CrossRef?Google Scholar??
42. iMoox Homepage. https://imoox.at/. Accessed 10 Feb 2022
43. Ingram, M., Reinschmidt, K., Schachter, K., Davidson, C., Sabo, S., Zapien, J., et al.: Establishing a professional profile of community health workers: results from a national study of roles, activities and training. J. Community Health 37, 529–37 (2011). https://doi.org/10.1007/s10900-011-9475-2
44. CrossRef?Google Scholar??
45. Izraeli, D., BarNir, A.: Promoting ethics through ethics officers: a proposed profile and an application. J. Bus. Ethics 17(11), 1189–1196 (1998). https://doi.org/10.1023/A:1005770732000
46. CrossRef?Google Scholar??
47. Janssen, M., Brous, P., Estevez, E., Barbosa, L.S., Janowski, T.: Data governance: organizing data for trustworthy artificial intelligence. Gov. Inf. Q. 37(3), 101493 (2020). https://doi.org/10.1016/j.giq.2020.101493
48. Kaptein, M.: Guidelines for the development of an ethics safety net. J. Bus. Ethics 41(3), 217–234 (2002). https://doi.org/10.1023/A:1021221211283
49. Kaptein, M.: Ethics programs and ethical culture: a next step in unraveling their multi-faceted relationship. J. Bus. Ethics 89, 261–281 (2008). https://doi.org/10.1007/s10551-008-9998-3
50. Kaptein, M.: The effectiveness of ethics programs: the role of scope, composition, and sequence. J. Bus. Ethics 132, 415–431 (2014). https://doi.org/10.1007/s10551-014-2296-3
51. Knowledge CheckR Homepage. https://www.knowledgecheckr.com/. Accessed 10 Feb 2022
52. Ladurner, C., Ortner, C., Lach, K., Ebner, M., Haas, M., Ebner, M., et al.: The development and implementation of missing tools and procedures at the interface of a university’s learning management system, its OER repository and the Austrian OER referatory. Int. J. Open Educ. Resour. 3(2), 25065 (2020)
53. Google Scholar??
54. Leitner, P., Ebner, M., Ammenwerth, E., Andergassen, M., Csanyi, G., Gr?blinger, O., et al.: Learning Analytics: Einsatz an ?sterreichischen Hochschulen. Workingpaper, Verein Forum neue Medien in der Lehre Austria, Austria (November 2019)
55. Google Scholar??
56. Leitner, P., Ebner, M., Geisswinkler, H., Sch?n, S.: Visualization of learning for students: a dashboard for study progress – development, design details, implementation, and user feedback. In: Sahin, M., Ifenthaler, D. (eds.) Visualizations and Dashboards for Learning Analytics. Advances in Analytics for Learning and Teaching. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-81222-5_19
57. Lima, R., Mesquita, D., Rocha, C., Rabelo, M.: Defining the industrial and engineering management professional profile: a longitudinal study based on job advertisements. Production 27, e20162299 (2017). https://doi.org/10.1590/0103-6513.229916
58. Llopis, J., Gonzalez, R., Gasco, J.: Corporate governance and organisational culture: the role of ethics officers. Int. J. Discl. Gov. 4, 96–105 (2007). https://doi.org/10.1057/palgrave.jdg.2050051
59. Mazur, T.C.: Ethics Officers, pp. 2028–2034. Springer, Heidelberg (2018). https://doi.org/10.1007/978-3-319-20928-9_2378
60. Morf, D.A., Schumacher, M.G., Vitell, S.J.: A survey of ethics officers in large organizations. J. Bus. Ethics 20(3), 265–271 (1999). https://doi.org/10.1023/A:1006000131803
61. Murr, T., Sch?n, S., Ebner, M.: By pupils for students: experience with the MOOC “Tenses Explained” (April 2021). https://doi.org/10.13140/RG.2.2.28346.11209
62. Pausits, A., Oppl, S., Sch?n, S., Fellner, M., Campell, D.: Distance Learning an ?sterreichischen Universit?ten und Hochschulen im Sommersemester 2020 und Wintersemester 2020/21. Bundesministerium für Bildung, Wissenschaft und Forschung (July 2021)
63. Google Scholar??
64. Remi?ová, A., La?áková, A., Kirchmayer, Z.: Influence of formal ethics program components on managerial ethical behavior. J. Bus. Ethics 160(1), 151–166 (2019). https://doi.org/10.1007/s10551-018-3832-3
65. Salda?a, D., Achury, L., Colmenares, S., Romero, H., Cavallo, E., Ulloa, A., et al.: Professional profile and work conditions of nurses working in intensive care units: a multicentre study. J. Clin. Nurs. (2021). https://doi.org/10.1111/jocn.16026
66. Scheidig, F., Holmeier, M.: Learning Analytics aus institutioneller Perspektive: Ein Orientierungsrahmen für die hochschulische Datennutzung, pp. 215–231 (October 2021). https://doi.org/10.1007/978-3-658-32849-8_13
67. Slade, S., Tait, A.: Global guidelines: ethics in learning analytics. Tech. rep. (March 2019). https://www.learntechlib.org/p/208251
68. Thomas, B.J., Hajiyev, J.: The direct and indirect effects of personality on data breach in education through the task-related compulsive technology use: M-learning perspective. IJCDS J. 9, 457–467 (2020). https://doi.org/10.12785/ijcds/090310
69. Trevi?o, L., den Nieuwenboer, N., Kreiner, G., Bishop, D.: Legitimating the legitimate: a grounded theory study of legitimacy work among ethics and compliance officers. Organ. Behav. Hum. Decis. Process. 123, 186–205 (2014). https://doi.org/10.1016/j.obhdp.2013.10.009
70. Vangarov, I.A., Belichev, M.G., Ilieva, N.I.: Application of statistical software SPSS to study the professional profile of engineers (nd)
71. Google Scholar??
72. Weber, J.: Implementing an organizational ethics program in an academic environment: the challenges and opportunities for the Duquesne university schools of business. J. Bus. Ethics 65(1), 23–42 (2006). https://doi.org/10.1007/s10551-005-3970-2
73. Wohinz, J.W.: Die Arbeitsplatzbeschreibung im Angestelltenbereich, pp. 1–8 (1976)
74. Google Scholar??
75. Zsolnai, L., Tencati, A.: The Future International Manager: A Vision of the Roles and Duties of Management. Palgrave Macmillan, London (2009)

76. Dimosthenis Mammonas "Artificial Intelligence Act: Council calls for promoting safe AI that respects fundamental rights" Council of the EU -Press release - 6 December 2022

77. European Commission "Proposal for a REGULATION OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL?- LAYING DOWN HARMONISED RULES ON ARTIFICIAL INTELLIGENCE (ARTIFICIAL INTELLIGENCE ACT) AND AMENDING CERTAIN UNION LEGISLATIVE ACTS" - EUR LEX 21.4.2021