View in EDS HTML Full Text PDF Full Text

Enhancing Ethical Codes with Artificial Intelligence Governance -- A Growing Necessity for the Adoption of Generative AI in Counselling

Saved in:
Bibliographic Details
Title: Enhancing Ethical Codes with Artificial Intelligence Governance -- A Growing Necessity for the Adoption of Generative AI in Counselling
Language: English
Authors: Pei Boon Ooi (ORCID 0000-0002-7259-3381), Graeme Wilkinson (ORCID 0000-0003-0709-222X)
Source: British Journal of Guidance & Counselling. 2025 53(1):66-80.
Availability: Routledge. Available from: Taylor & Francis, Ltd. 530 Walnut Street Suite 850, Philadelphia, PA 19106. Tel: 800-354-1420; Tel: 215-625-8900; Fax: 215-207-0050; Web site: http://www.tandf.co.uk/journals
Peer Reviewed: Y
Page Count: 15
Publication Date: 2025
Document Type: Journal Articles
Reports - Descriptive
Descriptors: Ethics, Artificial Intelligence, Governance, Computer Mediated Communication, Counseling, Psychotherapy, Foreign Countries, Decision Making, Algorithms
Geographic Terms: United States, Australia, United Kingdom, Malaysia
DOI: 10.1080/03069885.2024.2373180
ISSN: 0306-9885
1469-3534
Abstract: The advent of generative Artificial Intelligence (AI) systems, such as large language model chatbots, is likely to have a significant impact in psychotherapy and counselling in the future. In this paper we consider the current state of AI in psychotherapy and counselling and the likely evolution of this field. We examine the ethical codes of practice for counselling in four countries in different parts of the world, namely the UK, the USA, Australia and Malaysia, and identify aspects of these codes that will need enhancement to reflect good AI governance. Using the Model Artificial Intelligence Governance Framework as an example, we have identified how the key elements of the AI framework relate to the core elements of the ethical codes, as a pointer to how such ethical codes will need to be enhanced if generative AI systems are to be adopted by the counselling profession.
Abstractor: As Provided
Entry Date: 2025
Accession Number: EJ1473691
Database: ERIC
Full text is not displayed to guests.
FullText Links:
  – Type: pdflink
    Url: https://content.ebscohost.com/cds/retrieve?content=AQICAHj0k_4E0hTGH8RJwT4gCJyBsGNe_WN95AvKlDbXJGqwxwF35nfO2QbcSZB_FmQP9JqvAAAA4jCB3wYJKoZIhvcNAQcGoIHRMIHOAgEAMIHIBgkqhkiG9w0BBwEwHgYJYIZIAWUDBAEuMBEEDO5hw51CCla8B3BEfAIBEICBmklHOm0er470GD-qjhEUVZzY8ZtEHL9OE02jMWexmFruWT4jCcQHdP1vowlFWfj7ggOZzge-v2JEIvDR4qjlfUzYPalEAouvy0aWwSr2pmZYPyORdlV-eOUpIhwurs1ydvcTjk7uU9l0yJ7SBENHJzDsTGQNE4cVbOLIYU99ZBnnFPHkiwoFnE84KL-2pke0kEBv-2sKGW75Baw=
Text:
  Availability: 1
  Value: <anid>AN0185527847;bjg01feb.25;2025Jun02.03:40;v2.2.500</anid> <title id="AN0185527847-1">Enhancing ethical codes with artificial intelligence governance – a growing necessity for the adoption of generative AI in counselling </title> <p>The advent of generative Artificial Intelligence (AI) systems, such as large language model chatbots, is likely to have a significant impact in psychotherapy and counselling in the future. In this paper we consider the current state of AI in psychotherapy and counselling and the likely evolution of this field. We examine the ethical codes of practice for counselling in four countries in different parts of the world, namely the UK, the USA, Australia and Malaysia, and identify aspects of these codes that will need enhancement to reflect good AI governance. Using the Model Artificial Intelligence Governance Framework as an example, we have identified how the key elements of the AI framework relate to the core elements of the ethical codes, as a pointer to how such ethical codes will need to be enhanced if generative AI systems are to be adopted by the counselling profession.</p> <p>Keywords: SDG 3: Good health and well-being; AI in counselling; ethical codes; AI governance; psychotherapy</p> <hd id="AN0185527847-2">Introduction</hd> <p>In recent years there has been increasing recognition of the potential of computer-based or computer-supported mental health therapies – i.e. the utilisation of computers for providing mental health treatment or services directly to patients through interaction with computer programmes or delivered over the Internet (Carroll & Rounsaville, [<reflink idref="bib18" id="ref1">18</reflink>]) – including counselling and psychotherapy. Computer-based or computer-supported approaches can help to (i) bridge the gap between the growing demand for therapy or counselling and the limited supply of therapists' or counsellors' time, (ii) provide better access to mental health support for remote communities, and (iii) facilitate better engagement with certain potential client groups such as teenagers or students. Computer-based or computer-supported therapy or counselling can range from videoconferencing (Simpson, [<reflink idref="bib68" id="ref2">68</reflink>]) and online video sessions with psychotherapists or counsellors (Smith et al., [<reflink idref="bib69" id="ref3">69</reflink>]), which became commonplace during the Covid 19 pandemic, to use of online web-based systems or mobile applications ("apps") which either support face-to-face therapy or counselling sessions or even replace such sessions completely.</p> <p>There have been a number of studies on the effectiveness of computer-based mental health therapies particularly in regard to the delivery of cognitive behavioural therapy (CBT) (Finazzi et al., [<reflink idref="bib31" id="ref4">31</reflink>]; Huguet et al., [<reflink idref="bib40" id="ref5">40</reflink>]) especially in relation to the treatment of anxiety disorder and depression (e.g. Marshall et al., [<reflink idref="bib48" id="ref6">48</reflink>]), and on the effectiveness of online therapy during Covid 19 (e.g. Barker & Barker, [<reflink idref="bib9" id="ref7">9</reflink>]). While web-based and mobile app-based therapies can help fulfil unmet needs and appeal more to certain categories of potential clients, such as young people who are reluctant to engage in traditional face-to-face counselling (Paterson et al., [<reflink idref="bib55" id="ref8">55</reflink>]), there remain a number of issues and challenges associated with them such as concerns about data privacy, lack of underpinning theoretical frameworks, limited evidence of effectiveness, poor rates of long-term client retention, and need for better training of counsellors as identified in a number of studies (e.g. Baumel et al., [<reflink idref="bib10" id="ref9">10</reflink>]; Marshall et al., [<reflink idref="bib48" id="ref10">48</reflink>]; Phillips et al., [<reflink idref="bib59" id="ref11">59</reflink>]; Weisel et al., [<reflink idref="bib76" id="ref12">76</reflink>]). Also, although it has been found that many apps appear to use scientific claims in their marketing on online app stores, few provide genuine high-quality evidence of their effectiveness (Larsen et al., [<reflink idref="bib41" id="ref13">41</reflink>]). Other current challenges in delivering computer-based therapies include ensuring that practitioners are not only well-versed in digital culture themselves (i.e. on how to deliver online services) but are also well able to prepare their clients to best safeguard their own interests and well-being and avoid becoming victims from using online services – e.g. from cyberbullying (Chan et al., [<reflink idref="bib21" id="ref14">21</reflink>]), self-trolling, and Munchausen by Proxy by Internet (Anthony, [<reflink idref="bib6" id="ref15">6</reflink>]) in which someone falsely claims that another person is ill through social media platforms as an attention seeking behaviour.</p> <p>Nevertheless, despite the challenges and hurdles, what is becoming increasingly evident is that computer-supported or computer-based mental health therapies will continue to grow as increasing numbers of apps are developed, their quality and capabilities improve and the demand for mental health support in society continues on an upward trend. It has also been recognised that artificial intelligence (AI) approaches appear to have much potential in the domain of computer-based therapies. Moreover, recent advances in the field of AI are likely to catalyse further innovation in computer-based therapies with the potential to make them more realistic, empathetic and effective. However, the use of AI also generates significant issues of an ethical nature which need to be addressed.</p> <p>In this paper, we propose how AI governance should be incorporated into ethical codes for the psychotherapy and counselling professions in order to give better safeguards for clients in the use both of innovative AI-driven mental health apps as well as of AI-driven systems that might offer background support to clients engaged in traditional face-to-face therapy. For this paper, the terms "counselling" and "psychotherapy" will be used interchangeably, "practitioner" is used to refer to counsellors, psychotherapists, and their clinical supervisors (Wright, [<reflink idref="bib77" id="ref16">77</reflink>]), and "client" for clients and clinical patients.</p> <hd id="AN0185527847-3">Recent developments in AI</hd> <p>Although research on AI has been ongoing for several decades, the recent development of so-called "generative" AI systems has been a major leap forward in terms of AI capability. Moreover, generative AI systems based on "large language models" (LLMs), such as ChatGPT developed by the OpenAI organisation and made available for public use towards the end of 2022, have demonstrated unprecedented levels of performance in terms of being able to understand user queries and provide realistic and meaningful responses in a chatbot context. ChatGPT's performance in some reasoning tasks has even been found to exceed human performance (Webb et al., [<reflink idref="bib75" id="ref17">75</reflink>]). Novel interactive uses have already been found for this technology. For example, training it with past social media conversations involving a deceased person enabled it to chat with bereaved relatives in the manner of the deceased as though he or she were still alive, aiding in the grieving process (Pearcy, [<reflink idref="bib57" id="ref18">57</reflink>]). Such uses of course raise significant ethical questions.</p> <p>ChatGPT's sophistication, and that of similar systems, derives principally from the innovative deep learning neural network software architectures on which it is based, the very large volume of data on which it has been trained from public web-based sources, its ability to learn (including from new information provided by its users), and the enormous computing power used to provide its responses to millions of concurrent users all over the world. Such systems are now widely perceived as potentially revolutionary in many areas of human endeavour, including business (e.g. Mondal et al., [<reflink idref="bib50" id="ref19">50</reflink>]; Shaji George et al., [<reflink idref="bib66" id="ref20">66</reflink>]), education (e.g. AlAfnan et al., [<reflink idref="bib3" id="ref21">3</reflink>]; Cooper, [<reflink idref="bib23" id="ref22">23</reflink>]; Lim et al., [<reflink idref="bib44" id="ref23">44</reflink>]) and healthcare (e.g. Cascella et al., [<reflink idref="bib19" id="ref24">19</reflink>]; Chow et al., [<reflink idref="bib22" id="ref25">22</reflink>]) with potential application in medical education (Eysenbach, [<reflink idref="bib30" id="ref26">30</reflink>]).</p> <p>Although there are many perceived benefits to come from the wider adoption of AI, there remain a number of concerns such as that AI systems might one day become "superintelligent" (i.e. more intelligent than humans) and pose an existential risk to humanity (Bostrum, [<reflink idref="bib14" id="ref27">14</reflink>]; Cellan-Jones, [<reflink idref="bib20" id="ref28">20</reflink>]; Russell, [<reflink idref="bib62" id="ref29">62</reflink>]). Interestingly, however, diverse views on this emerged during 2023 with separate groups of scientists and experts writing open letters urging on the one hand caution in the further development of AI (Hern, [<reflink idref="bib39" id="ref30">39</reflink>]; Strickland & Zorpette, [<reflink idref="bib71" id="ref31">71</reflink>]) because of potential risks to humanity, and on the other hand urging for AI to be recognised as a force for good (British Computer Society, [<reflink idref="bib17" id="ref32">17</reflink>]). Regardless of which view is the more correct one, AI is likely to have a growing and profound impact on society in the coming years in that it is likely to replace tasks undertaken by professional workers in many industries with the potential for white-collar jobs to disappear (DePillis & Lohr, [<reflink idref="bib28" id="ref33">28</reflink>]). It has even been argued recently that the rapid pace of developments in AI will lead to a fifth industrial revolution, with a wave of innovations that may present significant risks and be uncontrollable (Suleyman & Bhaskar, [<reflink idref="bib72" id="ref34">72</reflink>]).</p> <p>Apart from the broader impact on professions and jobs and broader risks to society, AI systems may impact on individuals through the potential for false information or deepfakes to be easily generated and propagated with potential harm to individuals and democratic processes (Schick, [<reflink idref="bib63" id="ref35">63</reflink>]). Moreover, if AI systems become extensively used for business purposes they may lead to marginalisation of certain groups in society because of inherent biases arising from pre-existing human prejudices embedded in their training data (Nowotny, [<reflink idref="bib53" id="ref36">53</reflink>]). Such factors may generate a greater need for mental health services in the future, for example if instances of cyber-bullying and digital marginalisation increase significantly through the widespread use of AI. Further concerns about LLMs arise from the fact that although they can give responses that are often indistinguishable from outputs produced by a human, they lack cognitive ability and, although they are powerful, they represent a separation of agency and intelligence (Floridi, [<reflink idref="bib32" id="ref37">32</reflink>]). Therefore, patients or clients interacting with an LLM may think they are interacting with a system that has human-like intelligence when in fact they are not.</p> <hd id="AN0185527847-4">AI in psychotherapy and counselling – the situation to date</hd> <p>The potential of AI in mental health interventions, including psychotherapy and counselling has been increasingly recognised over the past decade. Some researchers believe AI will form the basis of a radical transformation or a new era in mental health care with the potential for precision or personalised mental health services and interventions (e.g. Alwadi & Lathifa, [<reflink idref="bib4" id="ref38">4</reflink>]; Bickman, [<reflink idref="bib11" id="ref39">11</reflink>]; Zhou et al., [<reflink idref="bib78" id="ref40">78</reflink>]). One view of the potential implementation of AI in counselling identified three pillars of counselling as: (a) forming a professional relationship, (b) empowering, and (c) accomplishing goals, with the conclusion that if one or two of these pillars could be met by an AI system it would be getting close to functioning as a counsellor (Fulmer, [<reflink idref="bib33" id="ref41">33</reflink>]). It appears that the latest LLMs are getting near this – a systematic review (Lim et al., [<reflink idref="bib45" id="ref42">45</reflink>]) revealed that chatbot-delivered psychotherapy was effective in meeting the needs of improving depressive symptoms among 1099 patients in 11 clinical trials.</p> <p>There have been increasing numbers of web-based systems and mobile apps being developed or prototyped that claim some use of AI approaches. Some of these systems have been evaluated with clinical trials, though mostly not. Normally these systems have been focused on particular types or purposes of therapy with specific target client groups, such as supporting youths recovering from psychosis (D'Alfonso et al., [<reflink idref="bib26" id="ref43">26</reflink>]), and relieving symptoms of depression and anxiety for students (Fulmer et al., [<reflink idref="bib34" id="ref44">34</reflink>]). AI has been proposed for preventive early treatment of mental health conditions and initial screening of clients (e.g. Li et al., [<reflink idref="bib43" id="ref45">43</reflink>]). Virtual counselling using AI has also been used to improve the communication skills of healthcare providers (Shorey et al., [<reflink idref="bib67" id="ref46">67</reflink>]). AI is also seen as a potential tool to evaluate the effectiveness of evidence-based psychotherapies through analysis of quality metrics from therapy sessions (Creed et al., [<reflink idref="bib25" id="ref47">25</reflink>]).</p> <p>There have been studies of the attitudes of mental health professionals towards the use of AI revealing some concerns about the lack of empathy of AI systems, data privacy, and potential loss of jobs (Blease et al., [<reflink idref="bib13" id="ref48">13</reflink>]), as well as the finding that those professionals involved in cognitive behavioural approaches appear more receptive to the idea of AI assisted therapies than those involved in psychodynamic approaches (Sebri et al., [<reflink idref="bib65" id="ref49">65</reflink>]). There have also been a number of studies of users' views of mental health apps and services using AI (e.g. Aktan et al., [<reflink idref="bib2" id="ref50">2</reflink>]; Götzl et al., [<reflink idref="bib36" id="ref51">36</reflink>]; Ma et al., [<reflink idref="bib47" id="ref52">47</reflink>]) revealing a number of findings such as a preference amongst some clients for labels such as mental "fitness" apps rather than mental "health" apps, general concern about privacy of data stored by systems, and a feeling that human therapists can be trusted more. Research has also shown a need for a clearer understanding of the place of AI in regard to the clinician-patient relationship and the need to define workflows for interactions which might involve AI systems, clinicians, patients, and administrators (Miner et al., [<reflink idref="bib49" id="ref53">49</reflink>]). A number of reviews have concluded that AI chatbots have the potential to reduce mental healthcare disparities and to deliver therapeutic techniques to assist clients with anxiety and depression and to provide psycho-education, but concerns over privacy, efficacy, and security remain, with the cautionary view being expressed that the field should avoid techno-fundamentalism (Gamble, [<reflink idref="bib35" id="ref54">35</reflink>]) which is essentially the idea that technology and innovation ultimately can solve all human problems. Studies have also shown that AI chatbots can make therapies scalable to large user bases and can sometimes be viewed as less judgmental than human therapists, though AI systems can sometimes appear repetitive in their interactions, misunderstand users, or give inappropriate responses (Boucher et al., [<reflink idref="bib15" id="ref55">15</reflink>]).</p> <p>Besides positive views on the potential of AI in therapy, negative views and concerns have also been expressed. Concern about equity is one of the most common, with the fear that AI models might perpetuate existing inequities in society regarding who gets diagnosed and treated and whether such treatment is effective (Timmons et al., [<reflink idref="bib74" id="ref56">74</reflink>]). A study in India in 2020 concluded that "the time is not yet ripe for the use of telepsychiatry or for AI to be adopted fully in the field of mental health care" (Thenral & Annamalai, [<reflink idref="bib73" id="ref57">73</reflink>], p. 432).</p> <p>The advent of LLM chatbots such as ChatGPT in late 2022 with their much-enhanced realism compared to earlier AI driven systems and their potential for improved interactional capability has given new impetus to the idea of AI driven therapy, though with an appreciation of the enhanced possibility for harmful feedback if not well regulated (Reardon, [<reflink idref="bib61" id="ref58">61</reflink>]). The potential uses of LLMs in psychotherapy have been described as lying on a set of stages ranging from purely assistive, to collaborative, to being fully autonomous (Stade et al., [<reflink idref="bib70" id="ref59">70</reflink>]). However, the over-arching need for elimination of risks has become a key issue identified in several studies (Hamdoun et al., [<reflink idref="bib38" id="ref60">38</reflink>]). This points to an urgent need to integrate concepts from good AI governance into the ethical codes of the psychotherapy and counselling professions, which we shall focus on in the following sections of this paper.</p> <hd id="AN0185527847-5">Features of psychotherapy and counselling codes of ethics</hd> <p>In considering how AI governance needs to be incorporated into psychotherapy and counselling practice, it is necessary to identify the key features that appear in the ethical codes or codes of practice adopted by the professional psychotherapy and counselling bodies around the world. There are of course many national variations, but most ethical codes or codes of practice have similar requirements and core components, which we have identified from codes of practice used in the UK, the USA, Australia, and Malaysia (American Counseling Association, [<reflink idref="bib5" id="ref61">5</reflink>]; Australian Counselling Association, [<reflink idref="bib8" id="ref62">8</reflink>]; British Association for The Counselling Professions, [<reflink idref="bib16" id="ref63">16</reflink>]; Lembaga Kaunselor Malaysia, [<reflink idref="bib42" id="ref64">42</reflink>]). What is apparent, however, is that none of the codes we have seen to date incorporate any special elements to take account of the rapid development and use of AI in the mental health field, which appears to be much needed. All ethical codes appear to cover the core and fundamental ethical principles of counselling which are autonomy, justice, beneficence, non-maleficence, fidelity or being trustworthy, and veracity. The four ethical codes for counselling or psychotherapy in current use that we have considered (UK, USA, Australia, and Malaysia) all essentially cover key issues that relate to those core principles such as: establishing an appropriate professional relationship and clear boundaries; respecting the client and ensuring the well-being and welfare of the client and the client's autonomy at all times; acting with integrity and avoiding harm to clients; maintaining appropriate documentation but always assuring confidentiality and privacy; and ensuring that practitioners have appropriate competence, training and supervision. Other matters are covered such as the need for building relevant connections with other professionals e.g. in primary care teams, the requirement always to act without bias or any form of discrimination, ensuring ethical issues are appropriately managed in relation to research and publication, and having appropriate mechanisms for handling complaints. It is not our purpose to compare lengthy ethical codes in detail here, suffice to say that there is much commonality, as might be expected for codes that relate to a relatively mature though evolving profession with much connectivity and sharing of knowledge and experience between professionals in different countries.</p> <p>Table 1 summarises the core elements of each of the four ethical codes we have considered from around the world.</p> <p>Table 1. Core elements of ethical codes of practice for counselling of four countries.</p> <p> <ephtml> <table><thead valign="bottom"><tr><td>Core Principles:</td><td>UK (2018)<xref ref-type="table-fn" rid="tfn2">1</xref></td><td>USA (2014)<xref ref-type="table-fn" rid="tfn3">2</xref></td><td>Australia (2021)<xref ref-type="table-fn" rid="tfn4">3</xref></td><td>Malaysia (2019)<xref ref-type="table-fn" rid="tfn5">4</xref></td></tr></thead><tbody><tr><td>Autonomy Justice Beneficence Non-Maleficence Fidelity/Being Trustworthy Veracity</td><td>Putting clients first <sup>D</sup></td><td>Counselling relationship <sup>B, D</sup></td><td>Issues of responsibility <sup>D</sup></td><td>Counselling relationship <sup>B, D</sup></td></tr><tr><td>Working with professional standards</td><td>Confidentiality and privacy <sup>D</sup></td><td>Responsibility to the clients <sup>B, D</sup></td><td>Confidentiality, privileged communication, and privacy <sup>D</sup></td></tr><tr><td>Respect</td><td>Professional responsibility</td><td>Counsellors' obligations on social media <sup>D</sup></td><td>Professional Responsibility <sup>B</sup></td></tr><tr><td>Building an appropriate relationship <sup>B, D</sup></td><td>Relationship with other professionals <sup>D</sup></td><td>Anti-discrimination practice</td><td>Relations with other professionals <sup>D</sup></td></tr><tr><td>Breaks and endings <sup>D</sup></td><td>Evaluation, assessment and interpretation <sup>D</sup></td><td>Confidentiality <sup>D</sup></td><td>Evaluation, assessment, and interpretation <sup>D</sup></td></tr><tr><td>Integrity <sup>C</sup></td><td>Supervision, training and teaching <sup>A, C</sup></td><td>Settings</td><td>Supervision, training, and teaching <sup>A, C</sup></td></tr><tr><td>Accountability and candour <sup>B</sup></td><td>Research and publication <sup>A, C</sup></td><td>Exceptional circumstances</td><td>Research and publication <sup>A, C</sup></td></tr><tr><td>Confidentiality <sup>D</sup></td><td>Distance counselling, technology and social media <sup>C</sup></td><td>Management and confidentiality</td><td>Resolving ethical issues <sup>A</sup></td></tr><tr><td>Working with colleagues in teams <sup>D</sup></td><td>Resolving ethical issues <sup>A</sup></td><td>Contracts <sup>C</sup></td><td /></tr><tr><td>Supervision <sup>A, C</sup></td><td /><td>Boundaries <sup>B, D</sup></td><td /></tr><tr><td>Training and education</td><td /><td>Dual and multiple roles <sup>B, D</sup></td><td /></tr><tr><td>Trainees</td><td /><td>Competence</td><td /></tr><tr><td>Research <sup>A, C</sup></td><td /><td>The counselling environment</td><td /></tr><tr><td>Care of self as a practitioner</td><td /><td>Committees <sup>A</sup></td><td /></tr><tr><td>Responding to ethical dilemmas & issues <sup>A</sup></td><td /><td /><td /></tr></tbody></table> </ephtml> </p> <p>1 Note: Superscript letters (A, B, C, D) refer to the top-level Model AI Governance Framework principles in Table 2.</p> <ulist> <item>2 (British Association for The Counselling Professions, [<reflink idref="bib16" id="ref65">16</reflink>]).</item> <item>3 (American Counseling Association, [<reflink idref="bib5" id="ref66">5</reflink>]).</item> <item>4 (Australian Counselling Association, [<reflink idref="bib8" id="ref67">8</reflink>]).</item> <item>5 (Lembaga Kaunselor Malaysia, [<reflink idref="bib42" id="ref68">42</reflink>]).</item> </ulist> <p>As mentioned earlier, none of these codes specifically mention AI; although two specifically mention technological issues, namely (i) the American code (which includes distance counselling and use of technology and complexities surrounding that such as the additional risks of technology assisted services including IT security, and potential challenges of operating across different geographical locations and legal jurisdictions), and (ii) the Australian code (which covers counsellors' obligations on social media and websites and the need specifically to not mis-represent their services or professional standing, amongst other related issues). The key issue now is how AI governance can be integrated with existing codes of ethics in order to produce updated codes which are appropriate for the emerging context of AI based counselling or psychotherapy.</p> <hd id="AN0185527847-6">Features of AI governance codes</hd> <p>The need to control AI or intelligent machines and ensure they are used responsibly and not counter to human interests has long been recognised. As early as the 1950s the science fiction writer Isaac Asimov created his three laws of robotics, the first of which states that a robot "may not injure a human being or, through inaction, allow a human being to come to harm" (Asimov, [<reflink idref="bib7" id="ref69">7</reflink>]). Such a law should certainly apply to any AI system today as a fundamental requirement, though contemporary AI systems need much more than this especially if used in the mental health field. In recent years, studies have focused on the need for AI systems to be loyal to their stakeholders (Aguirre et al., [<reflink idref="bib1" id="ref70">1</reflink>]), responsible (Peters et al., [<reflink idref="bib58" id="ref71">58</reflink>]), designed and used ethically (Luxton, [<reflink idref="bib46" id="ref72">46</reflink>]), and both fair and non-discriminatory and worthy of public trust (Cortes et al., [<reflink idref="bib24" id="ref73">24</reflink>]).</p> <p>With the rapid advances in AI in recent years, and especially the introduction of LLMs, there has been a growing recognition by governments worldwide of an urgent need to control AI development and use, partly owing to the concern about existential threats to humanity as mentioned earlier, but mainly because of the recognition that whilst AI can potentially yield enormous benefits to business and society there are significant risks and ethical issues associated with it. There has been a growing volume of research on AI governance and the establishment of research units and think tanks devoted to this topic such as the Centre for the Governance of AI, an independent non-profit organisation established in the UK in 2018.</p> <p>A number of governments have issued voluntary codes or frameworks regarding the ethical and safe use of AI with some working towards formal regulation or at least stating a need for legislation. The situation is highly dynamic with announcements of new policies and frameworks gathering pace. The Singapore government launched the first edition of its Model AI Governance Framework at the World Economic Forum in Davos in January 2019 and subsequently revised it (PDPC, [<reflink idref="bib56" id="ref74">56</reflink>]). Later in November 2019 the Australian government also issued its AI ethics framework (Department of Industry, [<reflink idref="bib27" id="ref75">27</reflink>]). In the United States a voluntary framework on AI risk management was issued early in 2023 (National Institute of Standards and Technology, [<reflink idref="bib52" id="ref76">52</reflink>]). President Biden later issued an Executive Order in October 2023 directing that actions should be taken on the safe, secure and trustworthy use of artificial intelligence which will require actions by software developers and the information technology industry, as well as by government agencies (Biden, [<reflink idref="bib12" id="ref77">12</reflink>]). In July 2023 the Cyberspace Administration of China also issued interim measures for the management of generative AI services aiming to encourage generative AI development but at the same time encouraging self-regulation by service providers without which services would be terminated (PWC China, [<reflink idref="bib60" id="ref78">60</reflink>]). The Canadian government also issued a voluntary code on the responsible development and management of generative AI in September 2023 (Government of Canada, [<reflink idref="bib37" id="ref79">37</reflink>]) and stated its intention to develop an Artificial Intelligence and Data Act. The UK government issued a policy paper on AI in August 2023 stating that it will develop a pro-innovation AI framework (Office for Artificial Intelligence, [<reflink idref="bib54" id="ref80">54</reflink>]), but has indicated it is not intending to regulate the British AI sector in the short term (Mosolova, [<reflink idref="bib51" id="ref81">51</reflink>]). The British position seems to crystallize more clearly than some others that policies are needed to ensure the safe and ethical use of AI without jumping too fast into regulation that might stifle innovation at a time of great technological change. The European Union has proposed a draft EU Artificial Intelligence Act that will focus particularly on high-risk uses of AI that might pose potential risk to the health, safety or fundamental rights of individuals (European Commission, [<reflink idref="bib29" id="ref82">29</reflink>]). It is likely this will become the first comprehensive legal framework for AI globally. The use of AI in counselling and psychotherapy would fall under the EU definition of "high-risk" usage of AI, because of the potential risk to human health and would then be subject to necessary legally enforceable safeguards once the EU AI Act is formally adopted.</p> <p>Of all of the AI governance frameworks issued so far, the one issued by the Singapore government is arguably one of the most comprehensive and appears to be the earliest to have been launched. We have therefore chosen to use it in this paper in considering how codes of ethics for counselling and psychotherapy should be enhanced to reflect the good principles of AI governance. The core elements of the Singapore government's Model AI Governance Framework (PDPC, [<reflink idref="bib56" id="ref83">56</reflink>]) for the use of AI cover the following four aspects (A-D):</p> <hd id="AN0185527847-7">A. Internal governance structures and measures</hd> <p>This section relates to the overall governance of the use of AI systems by organisations and ensuring that any ethical issues and risks are adequately managed and appropriately audited. This may involve, for example, the establishment of AI ethics boards. The establishment of such boards is a complex exercise, normally with five key design decisions (Schuett et al., [<reflink idref="bib64" id="ref84">64</reflink>]) covering its responsibilities, legal structure, membership, decision making processes, and resources needed. In the context of psychotherapy and counselling the issue of who establishes such a board is an important consideration. If AI is used to support the work of small counselling practices it would appear to be most relevant for such a board to be established at the level of the professional body which accredits individual counsellors or practices. However, there would be significant challenges in such a board then overseeing the day to day use of AI systems and ensuring that ethical risks are not being generated. This issue needs further consideration and research. At the level of a practice or a counselling services centre it would be possible to establish a Governance Council to review the standard operation procedures (SOPs) and the implementation of AI applications in the therapeutic sessions. Responsibilities would need to be assigned to different stakeholders such as:</p> <p></p> <ulist> <item> Director/ Head of counselling services: To review the current SOPs, revise them, and ensure that the implementation of AI is aligned with the ethical standards. To also conduct periodical audits to ensure compliance.</item> <p></p> <item> Counsellors: To be trained and competent in the usage and implementation of AI.</item> <p></p> <item> Clients: To obtain a comprehensive understanding of the nature of services that involve AI and provide informed consent to the services obtained.</item> </ulist> <hd id="AN0185527847-8">B. Determining the level of human involvement in AI-augmented decision-making</hd> <p>This is a key issue that relates to how human oversight of any AI system would operate. This requires a decision on the objectives of the AI and the risks of using it and how they will be handled. If an AI system is making decisions about clients, for example in relation to what counselling approaches to adopt or about whether clients are at risk of harm and might be in need of referral to other specialists, it is imperative that boundaries around what the AI can and cannot do are established. Moreover, there needs to be clarity about when and how human intervention will take place. Also, there need to be clear audit trails of why decisions have been taken by the AI system so that explanations can be made to relevant stakeholders which may include clients as well as professionals and professional bodies. The Model AI Governance Framework foresees three different kinds of human involvement in AI decision-making which are: (i) <emph>human-in-the-loop</emph>, in which humans have full control and take all decisions, (ii) <emph>human-out-of-the-loop</emph>, in which an AI system has full control and takes all decisions, and (iii) <emph>human-over-the-loop</emph>, in which humans act in a supervisory role and can take control if an AI system encounters situations it cannot deal or are unexpected. We can consider the implications of the three types human involvement in relation to the intake process of counselling services as an example:</p> <p></p> <ulist> <item> <emph>human-in-the-loop</emph>, in which counsellors or psychotherapists together with the clients would have full control and make all decisions regarding the treatment goals and plans of the therapeutic process.</item> <p></p> <item> <emph>human-out-of-the-loop</emph>, in which an AI system could have full control and manage the entire intake process (e.g. collection of client information sheets, following up with session reminders or finding the best time that suits all parties).</item> <p></p> <item> <emph>human-over-the-loop</emph>, in which supervisions or training are in place, especially with novice counsellors who are new to the adaptation or implementation of AI system and may need timely and frequent input or guidance.</item> </ulist> <p>In our view, in psychotherapy and counselling it is unlikely that any practical adoption of AI could operate in a <emph>human-out-of-the-loop</emph> mode because the risks of leaving all decisions to the AI system would be too high. When clinical decisions are made, they should be made in consultation among the stakeholders as it is essential to establish accountability. A <emph>human-out-of-the-loop</emph> system complicates the assignment of responsibility. This can lead to legal and ethical challenges, particularly in cases of harm or unintended consequences. Responsibility may lie largely, if not exclusively, with the software developers who may not be qualified to take clinical decisions made by the system. Furthermore, even though AI systems are advancing rapidly, they can misinterpret information and act unpredictably, particularly in unfamiliar situations or contexts where information they receive is contradictory or confusing. This problem has occurred in other fields of AI application. Self-driving cars, for example, have been involved in fatal accidents through misinterpreting visual sensor information.</p> <hd id="AN0185527847-9">C. Operations management</hd> <p>This relates to ensuring that an AI system uses appropriate data and algorithms in its operations, that data are managed appropriately and safeguarded, and that data outputs from an AI system are accurate and appropriate to the context and clients. A particular concern is to ensure that data used to train AI systems are robust and free from errors or biases. Such concerns are well-recognised in the case of ChatGPT and other large language model AI systems that use massive quantities of data taken from the Internet given that information on the Internet may be biased and inaccurate. If large language model AI systems are used in psychotherapy and counselling there needs to be some control over the types of data sources they exploit. At the present time users of ChatGPT, for example, have no control over the data on which it has been trained.</p> <p>Counsellors may use LLMs to generate reference materials, exercises or homework for treatment purposes, but the outputs generated must be aligned with the treatment objectives and goals, and be appropriate for the particular clients. Additionally, counsellors must ensure high standards of data governance (e.g. who has access and how long will the records be available or accessible), data accuracy (e.g. are the materials generated culturally appropriate and meeting the needs of the therapy), and representativeness (e.g. are any materials created by the AI system generalisable to the current group of clients).</p> <p>Some counsellors may allow recording with the aim of letting clients use it as reference material or for homework. Thus, procedures for release of information will need to be incorporated into the operations management process. Other operations management issues also need to be addressed, such as during the intake process deciding on which platform to use, how passwords are to be protected or other security measures adopted, and what is needed from the clients before starting sessions (e.g. device and user compatibility). The operations management process should also refer to and maintain compliance with relevant laws, such as with the US Health Insurance Portability and Accountability Act (HIPAA) of 1996.</p> <hd id="AN0185527847-10">D. Stakeholder interaction and communication</hd> <p>This relates to the need for transparency and clear communication to stakeholders about if and how AI is to be used. In the context of psychotherapy and counselling absolute clarity is required on the purpose of the use of AI, how it will be used, potential risks, and on how clients can give feedback to humans involved in oversight of the AI, and if necessary on how request to opt-out of use of the AI. Computer-based or computer-supported therapy or counselling systems using AI must address the importance of respect for individuals, foster trust between counsellors and clients, ensure psychological well-being and uphold honest communication. The interaction and engagement with various stakeholders must be effective at all stages, including during pre-session activity (e.g. advertisement of the service should be clear on the use of the AI and on the 5W and 1H questions – what, who, why, where, when, and how – relating to the process), during the sessions (e.g. clarity on what the potential benefits or risks are and on rights to withdraw) and post-sessions (e.g. dealing with potential technological dependency and what are the resources or referral systems in place to support the clients).</p> <p>The main features of the Model AI Governance Framework are shown in Table 2.</p> <p>Table 2. Main elements of the Singapore Model AI Governance Framework.</p> <p> <ephtml> <table><tbody><tr><td><bold>A.</bold></td><td><bold>Internal Governance Structures and Measures</bold></td></tr><tr><td>Clear roles and responsibilities for the ethical deployment of AI Risk management and internal controls</td></tr><tr><td><bold>B</bold></td><td><bold>Determining the Level of Human Involvement in AI-Augmented Decision-Making</bold></td></tr><tr><td>Having clear objective in use of AI Human-in-the-loop, human-out-of-the-loop, human-over-the-loop decision-making Probability-severity assessment</td></tr><tr><td><bold>C.</bold></td><td><bold>Operations Management</bold></td></tr><tr><td>Good data accountability practices (lineage of data, data provenance records, ensuring data quality, minimising inherent biases etc) Ensuring transparency of system algorithms (explainability, repeatability, tuning of system, auditability)</td></tr><tr><td><bold>D.</bold></td><td><bold>Stakeholder Interaction and Communication</bold></td></tr><tr><td>Disclosure on why and how AI is used Explainability and transparency, opt-out options Testing the user interface, acceptable user policies</td></tr></tbody></table> </ephtml> </p> <p>If AI is to become widely used in psychotherapy and counselling, as seems likely, it is clear that the requirements of AI governance frameworks, such as that of Singapore, will need to be embedded within the ethical codes used by the profession. We briefly consider this in the next section.</p> <hd id="AN0185527847-11">Enhancing psychotherapy and counselling ethics codes with AI governance</hd> <p>Given the likely convergence of mental health app technology and generative AI technology, it seems inevitable that AI supported psychotherapy or counselling will become a reality in the near future. Precisely how AI systems might be used, in what contexts, and how widespread they will become still remains to be seen. However, it now seems necessary for ethical codes of practice relating to the profession to be enhanced with the elements of good AI governance. We have therefore undertaken a top-level analysis of which elements of AI governance are relevant to which aspects of the ethical codes of the four countries we have examined. The results of this analysis are shown in Table 3, which highlights the likely areas of intersection or interaction between the elements of the Model AI Governance Framework and the top-level sections of the four national ethical codes.</p> <p>Table 3. Analysis of probable areas of intersection of counselling/psychotherapy ethical code elements and the Model AI Governance Framework core elements (using UK, US, Australian, and Malaysian Counselling Codes and the Singapore Model AI Governance Framework).</p> <p> <ephtml> <table><tbody><tr><td><bold>A. Internal Governance Structures and Measures (Model AI Governance Framework)</bold></td></tr><tr><td /><td>British (2018)</td><td>USA (2014)</td><td>Australia (June 2022)</td><td>Malaysia (1998)</td></tr><tr><td>Elements of Ethical Codes:</td><td>Supervision</td><td>Supervision, training and teaching</td><td>Committees</td><td>Supervision, training, and teaching</td></tr><tr><td>Research</td><td>Resolving ethical issues</td><td /><td>Research and publication</td></tr><tr><td>Responding to ethical dilemma & issues</td><td /><td /><td>Resolving ethical issues</td></tr><tr><td><bold>B. Determining the Level of Human Involvement in AI-Augmented Decision-Making (Model AI Governance Framework)</bold></td></tr><tr><td /><td>British (2018)</td><td>USA (2014)</td><td>Australia (June 2022)</td><td>Malaysia (1998)</td></tr><tr><td>Elements of Ethical Codes:</td><td>Building an appropriate relationship</td><td>Counselling relationship</td><td>Responsibility to the clients</td><td>Counselling relationship</td></tr><tr><td>Accountability and candour</td><td /><td>Boundaries</td><td>Professional Responsibility</td></tr><tr><td><bold>C. Operations Management (Model AI Governance Framework)</bold></td></tr><tr><td /><td>British (2018)</td><td>USA (2014)</td><td>Australia (June 2022)</td><td>Malaysia (1998)</td></tr><tr><td>Elements of Ethical Codes:</td><td>Integrity</td><td>Supervision, training and teaching</td><td>Contracts</td><td>Supervision, training, and teaching</td></tr><tr><td>Supervision</td><td>Research and publication</td><td /><td>Research and publication</td></tr><tr><td>Research</td><td>Distance counselling, technology and social media</td><td /><td /></tr><tr><td><bold>D. Stakeholder Interaction and Communication (Model AI Governance Framework)</bold></td></tr><tr><td /><td>British (2018)</td><td>USA (2014)</td><td>Australia (June 2022)</td><td>Malaysia (1998)</td></tr><tr><td>Elements of Ethical Codes:</td><td>Putting clients first</td><td>Counselling relationship</td><td>Issues of responsibility</td><td>Counselling relationship</td></tr><tr><td>Building an appropriate relationship</td><td>Confidentiality and privacy</td><td>Responsibility to the clients</td><td>Confidentiality, privileged communication, and privacy</td></tr><tr><td>Breaks and endings</td><td>Relationships with other professionals</td><td>Counsellors' obligations on social media</td><td>Relations with other professionals</td></tr><tr><td>Confidentiality</td><td>Evaluation, assessment and interpretation</td><td>Confidentiality</td><td>Evaluation, assessment, and interpretation</td></tr><tr><td>Working with colleagues in teams</td><td /><td>Boundaries</td><td /></tr><tr><td /><td /><td>Dual and multiple roles</td><td /></tr></tbody></table> </ephtml> </p> <p>In our view, the sections of the ethical codes for the profession identified in the table should be enhanced with content reflecting the spirit and intention of the AI governance framework. This would be a complex task requiring the involvement of the leadership of the profession in different countries, and we do not presume to attempt such a significant task here. Our purpose is to highlight the need for this and to point out likely areas of the ethical codes in the four countries which will need enhancement.</p> <p>However, even without attempting to enhance a complete counselling ethical code with AI governance principles, we can give an indication of the kind of enhancement that would be needed by considering one element of the British ethical code and one area of the example Model AI Governance Framework from Singapore. Area B in the Model AI Governance Framework (Determining the Level of Human Involvement in AI-Augmented Decision-Making) focuses on the appropriate extent of human oversight with AI systems. This area is closely related to the fundamental principle of psychotherapy and counselling – i.e. the formation of working alliances and relationships between clients and counsellors. The section "Building an Appropriate Relationship" in the British Association for the Counselling Professions Ethical Framework for the Counselling Professions (British Association for The Counselling Professions, [<reflink idref="bib16" id="ref85">16</reflink>]), for example, outlines the criteria and guidelines for how a professional therapeutic relationship is established. Future revisions of this ethical code might consider stating the objectives, potential strengths, and risks of using AI systems in forming therapeutic relationships. A probability-severity assessment matrix could be used to help clients and psychotherapists determine the level of human involvement, minimise AI dependency, and avoid an AI system causing harm in any therapeutic relationship. The counselling agreement should state the objective of having an AI system in the therapeutic relationship, as compared to a traditional in-person client-counsellor relationship, and how the dynamic of the therapeutic relationship would change with an AI system. The ethical code would also make clear the need to communicate clearly to clients who interact with or are treated by an AI system the clients' and psychotherapists' roles in any decision-making processes, review potential risks periodically, and evaluate the effects at the termination stage with the co-existence of AI systems.</p> <p>The foregoing analysis cannot be considered definitive, nor without potential for much debate. Our discussion has been necessarily limited by the consideration of only one significant AI governance framework, i.e. the Model AI Governance Framework of Singapore as an illustration. As new AI governance frameworks are developed around the world, there may be individual nuances and requirements applicable to specific national contexts which will need to be reflected in the relevant psychotherapy and counselling ethical codes.</p> <hd id="AN0185527847-12">Conclusions</hd> <p>It is now becoming apparent that generative AI is likely to be exploited in many areas of human activity and it is unlikely that psychotherapy and counselling will be left out of that. Already mental health apps have found value in certain contexts, and the potential of AI chatbots to engage clients meaningfully in limited experimental situations has been demonstrated. The operational use of AI in mental health, and especially in psychotherapy and counselling seems inevitable in the near future. In our view the time is now right to re-examine the content of the ethical codes of the psychotherapy and counselling professions and to enhance them with the necessary principles of good AI governance. This is likely to be an important and non-trivial task, but it ought to be carried out before AI systems for psychotherapy or counselling become widespread otherwise there is a risk of such systems being deployed in an ethical vacuum.</p> <p>Much discussion is taking place internationally about the need for control and governance of AI and its application with governments expressing clear intentions to act especially in regard to high risk uses. A number of governments have developed voluntary AI ethical frameworks though few have so far taken steps towards compulsory frameworks or legislation. In this paper we have analysed how the core elements of four counselling ethical codes intersect with one national AI governance framework proposing that elements of AI governance must therefore be integrated into such codes. However, this is an evolving situation as ethical codes are regularly updated and more AI governance frameworks are being developed and released globally. It seems imperative that as new frameworks are developed they should be reviewed for their potential impact on counselling ethical codes. This is likely to create a requirement for regular reviews and refreshment of ethical codes to reflect the latest thinking on the issue, ideally with the sharing of ideas and best practices internationally, and with further analyses being conducted as documented in this paper. The development of a new generation of ethical codes to reflect good AI governance, which appears to be a necessity in a highly dynamic situation, is clearly a complex matter and one that is of major and increasing significance to the psychotherapy and counselling profession. It is closely interconnected with issues such as the professional acceptability of AI exploitation and it is therefore foreseen that any discussions about enhancements to ethical codes will need to be part of much broader debates about the role of AI in relation to the roles of professionals. It is hoped that this paper might catalyze such debates and discussions and stimulate the enhancement of ethical codes in the near future.</p> <hd id="AN0185527847-13">Data availability statement</hd> <p>Data sharing is not applicable to this article as no new data were created or analyzed in this study.</p> <hd id="AN0185527847-14">Declarations</hd> <p>No funding was received to assist with the preparation of this manuscript. The authors have no competing interests to declare that are relevant to the content of this article.</p> <hd id="AN0185527847-15">Disclosure statement</hd> <p>No potential conflict of interest was reported by the author(s).</p> <ref id="AN0185527847-16"> <title> References </title> <blist> <bibl id="bib1" idref="ref70" type="bt">1</bibl> <bibtext> Aguirre, A., Dempsey, G., Surden, H., & Reiner, P. B. (2020). AI loyalty: A new paradigm for aligning stakeholder interests. IEEE Transactions on Technology and Society, 1 (3), 128 – 137. https://doi.org/10.1109/TTS.2020.3013490</bibtext> </blist> <blist> <bibl id="bib2" idref="ref50" type="bt">2</bibl> <bibtext> Aktan, M. E., Turhan, Z., & Dolu, İ. (2022). Attitudes and perspectives towards the preferences for artificial intelligence in psychotherapy. Computers in Human Behavior, 133, 107273. https://doi.org/10.1016/j.chb.2022.107273</bibtext> </blist> <blist> <bibl id="bib3" idref="ref21" type="bt">3</bibl> <bibtext> AlAfnan, M. A., Dishari, S., Jovic, M., & Lomidze, K. (2023). ChatGPT as an educational tool: Opportunities, challenges, and recommendations for communication, business writing, and composition courses. Journal of Artificial Intelligence and Technology, https://doi.org/10.37965/JAIT.2023.0184</bibtext> </blist> <blist> <bibl id="bib4" idref="ref38" type="bt">4</bibl> <bibtext> Alwadi, A., & Lathifa, Z. (2022). Applications of artificial intelligence in the treatment of behavioral and mental health conditions. Applied Research in Artificial Intelligence and Cloud Computing, 5 (1), 1 – 18.</bibtext> </blist> <blist> <bibl id="bib5" idref="ref61" type="bt">5</bibl> <bibtext> American Counseling Association. (2014). 2014 ACA code of ethics. https://<ulink href="http://www.counseling.org/resources/aca-code-of-ethics.pdf">www.counseling.org/resources/aca-code-of-ethics.pdf</ulink></bibtext> </blist> <blist> <bibl id="bib6" idref="ref15" type="bt">6</bibl> <bibtext> Anthony, K. (2015). Training therapists to work effectively online and offline within digital culture. British Journal of Guidance & Counselling, 43 (1), 36 – 42. https://doi.org/10.1080/03069885.2014.924617</bibtext> </blist> <blist> <bibl id="bib7" idref="ref69" type="bt">7</bibl> <bibtext> Asimov, I. (1950). Runaround. In I, Robot (The Isaac Asimov Collection). Doubleday.</bibtext> </blist> <blist> <bibl id="bib8" idref="ref62" type="bt">8</bibl> <bibtext> Australian Counselling Association. (2022). Code of ethics and practice of the Australian counselling association. https://<ulink href="http://www.theaca.net.au/documents/ACA-Code-of-Ethics-and-Practice-Ver16.pdf">www.theaca.net.au/documents/ACA-Code-of-Ethics-and-Practice-Ver16.pdf</ulink></bibtext> </blist> <blist> <bibl id="bib9" idref="ref7" type="bt">9</bibl> <bibtext> Barker, G. G., & Barker, E. E. (2022). Online therapy: Lessons learned from the COVID-19 health crisis. British Journal of Guidance & Counselling, 50 (1), 66 – 81. https://doi.org/10.1080/03069885.2021.1889462</bibtext> </blist> <blist> <bibtext> Baumel, A., Muench, F., Edan, S., & Kane, J. M. (2019). Objective user engagement with mental health apps: Systematic search and panel-based usage analysis. Journal of Medical Internet Research, 21 (9), e14567. https://doi.org/10.2196/14567</bibtext> </blist> <blist> <bibtext> Bickman, L. (2020). Improving mental health services: A 50-year journey from randomized experiments to artificial intelligence and precision mental health. Administration and Policy in Mental Health and Mental Health Services Research, 47 (5), 795 – 843. https://doi.org/10.1007/s10488-020-01065-8</bibtext> </blist> <blist> <bibtext> Biden, J. R., Jr. (2023). Executive order on safe, secure, and trustworthy artificial intelligence. The White House. https://<ulink href="http://www.whitehouse.gov/briefing-room/presidential-actions/2023/10/30/executive-order-on-the-safe-secure-and-trustworthy-development-and-use-of-artificial-intelligence/">www.whitehouse.gov/briefing-room/presidential-actions/2023/10/30/executive-order-on-the-safe-secure-and-trustworthy-development-and-use-of-artificial-intelligence/</ulink></bibtext> </blist> <blist> <bibtext> Blease, C., Locher, C., Leon-Carlyle, M., & Doraiswamy, M. (2020). Artificial intelligence and the future of psychiatry: Qualitative findings from a global physician survey. Digital Health, 6, 205520762096835. https://doi.org/10.1177/2055207620968355</bibtext> </blist> <blist> <bibtext> Bostrum, N. (2014). Superintelligence: Paths, dangers, strategies. Oxford University Press.</bibtext> </blist> <blist> <bibtext> Boucher, E. M., Harake, N. R., Ward, H. E., Stoeckl, S. E., Vargas, J., Minkel, J., Parks, A. C., & Zilca, R. (2021). Artificially intelligent chatbots in digital mental health interventions: A review. Expert Review of Medical Devices, 18 (sup1), 37 – 49. https://doi.org/10.1080/17434440.2021.2013200</bibtext> </blist> <blist> <bibtext> British Association for The Counselling Professions. (2018). Ethical framework for the counselling professions. https://<ulink href="http://www.bacp.co.uk/media/3103/bacp-ethical-framework-for-the-counselling-professions-2018.pdf">www.bacp.co.uk/media/3103/bacp-ethical-framework-for-the-counselling-professions-2018.pdf</ulink>.</bibtext> </blist> <blist> <bibtext> British Computer Society. (2023, July 18). BCS open letter calls for AI to be recognised as "force for good not threat to humanity". https://<ulink href="http://Www.Bcs.Org/Articles-Opinion-and-Research/Bcs-Open-Letter-Calls-for-Ai-to-Be-Recognised-as-Force-for-Good-Not-Threat-to-Humanity/">Www.Bcs.Org/Articles-Opinion-and-Research/Bcs-Open-Letter-Calls-for-Ai-to-Be-Recognised-as-Force-for-Good-Not-Threat-to-Humanity/</ulink></bibtext> </blist> <blist> <bibtext> Carroll, K. M., & Rounsaville, B. J. (2010). Computer-assisted therapy in psychiatry: Be brave – It's a new world. Current Psychiatry Reports, 12 (5), 426 – 432. https://doi.org/10.1007/s11920-010-0146-2</bibtext> </blist> <blist> <bibtext> Cascella, M., Montomoli, J., Bellini, V., & Bignami, E. (2023). Evaluating the feasibility of ChatGPT in healthcare: An analysis of multiple clinical and research scenarios. Journal of Medical Systems, 47 (33), https://doi.org/10.1007/S10916-023-01925-4</bibtext> </blist> <blist> <bibtext> Cellan-Jones, R. (2014, December 2). Stephen Hawking warns artificial intelligence could end mankind. BBC News : https://<ulink href="http://www.Bbc.Com/News/Technology-30290540">www.Bbc.Com/News/Technology-30290540</ulink></bibtext> </blist> <blist> <bibtext> Chan, N. N., Ahrumugam, P., Scheithauer, H., Schultze-Krumbholz, A., & Ooi, P. B. (2020). A hermeneutic phenomenological study of students' and school counsellors' "lived experiences" of cyberbullying and bullying. Computers & Education, 146, 103755. https://doi.org/10.1016/j.compedu.2019.103755</bibtext> </blist> <blist> <bibtext> Chow, J. C. L., Sanders, L., & Li, K. (2023). Impact of ChatGPT on medical chatbots as a disruptive technology. Frontiers in Artificial Intelligence, 6. https://doi.org/10.3389/FRAI.2023.1166014</bibtext> </blist> <blist> <bibtext> Cooper, G. (2023). Examining science education in ChatGPT: An exploratory study of generative artificial intelligence. Journal of Science Education and Technology, 32 (3), 444 – 452. https://doi.org/10.1007/S10956-023-10039-Y</bibtext> </blist> <blist> <bibtext> Cortes, U., Cortes, A., Barrue, C., Sanchez, A., Moya-Sanchez, E. U., & Garcia-Gasulla, D. (2021). To be fAIr or not to be: Using AI for the good of citizens. IEEE Technology and Society Magazine, 40 (1), 55 – 70. https://doi.org/10.1109/MTS.2021.3056173</bibtext> </blist> <blist> <bibtext> Creed, T. A., Salama, L., Slevin, R., Tanana, M., Imel, Z., Narayanan, S., & Atkins, D. C. (2022). Enhancing the quality of cognitive behavioral therapy in community mental health through artificial intelligence generated fidelity feedback (Project AFFECT): A study protocol. BMC Health Services Research, 22 (1), 1177. https://doi.org/10.1186/s12913-022-08519-9</bibtext> </blist> <blist> <bibtext> D'Alfonso, S., Santesteban-Echarri, O., Rice, S., Wadley, G., Lederman, R., Miles, C., Gleeson, J., & Alvarez-Jimenez, M. (2017). Artificial intelligence-assisted online social therapy for youth mental health. Frontiers in Psychology, 8. https://doi.org/10.3389/fpsyg.2017.00796</bibtext> </blist> <blist> <bibtext> Department of Industry, S. and R. (2019). Australia's artificial intelligence ethics framework. https://<ulink href="http://www.industry.gov.au/publications/australias-artificial-intelligence-ethics-framework">www.industry.gov.au/publications/australias-artificial-intelligence-ethics-framework</ulink></bibtext> </blist> <blist> <bibtext> DePillis, L., & Lohr, S. (2023). Tinkering with ChatGPT, workers wonder: Will this take my job? The New York Times. https://<ulink href="http://www.nytimes.com/2023/03/28/business/economy/jobs-ai-artificial-intelligence-chatgpt.html">www.nytimes.com/2023/03/28/business/economy/jobs-ai-artificial-intelligence-chatgpt.html</ulink></bibtext> </blist> <blist> <bibtext> European Commission. (2021). 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. https://eur-lex.europa.eu/resource.html?uri=cellar:e0649735-a372-11eb-9585-01aa75ed71a1.0001.02/DOC_1&format=PDF</bibtext> </blist> <blist> <bibtext> Eysenbach, G. (2023). The role of ChatGPT, generative language models, and artificial intelligence in medical education: A conversation with ChatGPT and a call for papers. JMIR Medical Education, 9, e46885. https://doi.org/10.2196/46885</bibtext> </blist> <blist> <bibtext> Finazzi, E., MacLeod, E., & MacBeth, A. (2023). Exploring service users experiences of remotely delivered CBT interventions in primary care during COVID-19: An interpretative phenomenological analysis. PLoS One, 18 (1), e0279263. https://doi.org/10.1371/journal.pone.0279263</bibtext> </blist> <blist> <bibtext> Floridi, L. (2023). AI as agency without intelligence: On ChatGPT, large language models, and other generative models. Philosophy and Technology, 36.</bibtext> </blist> <blist> <bibtext> Fulmer, R. (2019). Artificial intelligence and counseling: Four levels of implementation. Theory & Psychology, 29 (6), 807 – 819. https://doi.org/10.1177/0959354319853045</bibtext> </blist> <blist> <bibtext> Fulmer, R., Joerin, A., Gentile, B., Lakerink, L., & Rauws, M. (2018). Using psychological artificial intelligence (Tess) to relieve symptoms of depression and anxiety: Randomized controlled trial. JMIR Mental Health, 5 (4), e64. https://doi.org/10.2196/mental.9782</bibtext> </blist> <blist> <bibtext> Gamble, A. (2020). Artificial intelligence and mobile apps for mental healthcare: A social informatics perspective. Aslib Journal of Information Management, 72 (4), 509 – 523. https://doi.org/10.1108/AJIM-11-2019-0316</bibtext> </blist> <blist> <bibtext> Götzl, C., Hiller, S., Rauschenberg, C., Schick, A., Fechtelpeter, J., Fischer Abaigar, U., Koppe, G., Durstewitz, D., Reininghaus, U., & Krumm, S. (2022). Artificial intelligence-informed mobile mental health apps for young people: A mixed-methods approach on users' and stakeholders' perspectives. Child and Adolescent Psychiatry and Mental Health, 16 (1), 86. https://doi.org/10.1186/s13034-022-00522-6</bibtext> </blist> <blist> <bibtext> Government of Canada. (2023). Voluntary code of conduct on the responsible development and management of advanced generative AI systems. https://ised-isde.canada.ca/site/ised/en/voluntary-code-conduct-responsible-development-and-management-advanced-generative-ai-systems</bibtext> </blist> <blist> <bibtext> Hamdoun, S., Monteleone, R., Bookman, T., & Michael, K. (2023). AI-based and digital mental health apps: Balancing need and risk. IEEE Technology and Society Magazine, 42 (1), 25 – 36. https://doi.org/10.1109/MTS.2023.3241309</bibtext> </blist> <blist> <bibtext> Hern, A. (2023, March 29). Elon Musk joins call for pause in creation of giant AI "digital minds". The Guardian</bibtext> </blist> <blist> <bibtext> Huguet, A., Rao, S., McGrath, P. J., Wozney, L., Wheaton, M., Conrod, J., & Rozario, S. (2016). A systematic review of cognitive behavioral therapy and behavioral activation apps for depression. PLoS One, 11 (5), e0154248. https://doi.org/10.1371/journal.pone.0154248</bibtext> </blist> <blist> <bibtext> Larsen, M. E., Huckvale, K., Nicholas, J., Torous, J., Birrell, L., Li, E., & Reda, B. (2019). Using science to sell apps: Evaluation of mental health app store quality claims. Npj Digital Medicine, 2 (18). https://doi.org/10.1038/s41746-019-0093-1</bibtext> </blist> <blist> <bibtext> Lembaga Kaunselor Malaysia. (2019). Board of counsellors code of ethics.</bibtext> </blist> <blist> <bibtext> Li, Y., Deng, K., & Chen, X. (2020). The application of artificial intelligence in psychological counseling based on "treat pre-disease". Proceedings of the 2020 Conference on Artificial Intelligence and Healthcare, 72 – 78. https://doi.org/10.1145/3433996.3434010</bibtext> </blist> <blist> <bibtext> Lim, W. M., Gunasekara, A., Pallant, J. L., Pallant, J. I., & Pechenkina, E. (2023). Generative AI and the future of education: Ragnarök or reformation? A paradoxical perspective from management educators. The International Journal of Management Education, 21 (2), 100790. https://doi.org/10.1016/j.ijme.2023.100790</bibtext> </blist> <blist> <bibtext> Lim, S. M., Shiau, C. W. C., Cheng, L. J., & Lau, Y. (2022). Chatbot-delivered psychotherapy for adults with depressive and anxiety symptoms: A systematic review and meta-regression. Behavior Therapy, 53 (2), 334 – 347. https://doi.org/10.1016/j.beth.2021.09.007</bibtext> </blist> <blist> <bibtext> Luxton, D. D. (2014). Recommendations for the ethical use and design of artificial intelligent care providers. Artificial Intelligence in Medicine, 62 (1), 1 – 10. https://doi.org/10.1016/j.artmed.2014.06.004</bibtext> </blist> <blist> <bibtext> Ma, J. S., O'Riordan, M., Mazzer, K., Batterham, P. J., Bradford, S., Kõlves, K., Titov, N., Klein, B., & Rickwood, D. J. (2022). Consumer perspectives on the use of artificial intelligence technology and automation in crisis support services: Mixed methods study. JMIR Human Factors, 9 (3), e34514. https://doi.org/10.2196/34514</bibtext> </blist> <blist> <bibtext> Marshall, J. M., Dunstan, D. A., & Bartik, W. (2020). Apps with maps – Anxiety and depression mobile apps with evidence-based frameworks: Systematic search of major app stores. JMIR Mental Health, 7 (6), e16525. https://doi.org/10.2196/16525</bibtext> </blist> <blist> <bibtext> Miner, A. S., Shah, N., Bullock, K. D., Arnow, B. A., Bailenson, J., & Hancock, J. (2019). Key considerations for incorporating conversational AI in psychotherapy. Frontiers in Psychiatry, 10. https://doi.org/10.3389/fpsyt.2019.00746</bibtext> </blist> <blist> <bibtext> Mondal, S., Das, S., & Vrana, V. G. (2023). How to bell the cat? A theoretical review of generative artificial intelligence towards digital disruption in all walks of life. Technologies, 11 (2), 44. https://doi.org/10.3390/technologies11020044</bibtext> </blist> <blist> <bibtext> Mosolova, D. (2023). UK will refrain from regulating AI "in the short term". The Financial Times, November 17. https://<ulink href="http://www.ft.com/content/ecef269b-be57-4a52-8743-70da5b8d9a65">www.ft.com/content/ecef269b-be57-4a52-8743-70da5b8d9a65</ulink></bibtext> </blist> <blist> <bibtext> National Institute of Standards and Technology. (2023). Artificial intelligence risk management framework (AI RMF 1.0). https://nvlpubs.nist.gov/nistpubs/ai/NIST.AI.100-1.pdf</bibtext> </blist> <blist> <bibtext> Nowotny, H. (2021). In AI we trust – Power, illusion and control of predictive algorithms. Polity Press.</bibtext> </blist> <blist> <bibtext> Office for Artificial Intelligence. (2023). A pro-innovation approach to AI regulation. https://<ulink href="http://www.gov.uk/government/publications/ai-regulation-a-pro-innovation-approach/white-paper">www.gov.uk/government/publications/ai-regulation-a-pro-innovation-approach/white-paper</ulink></bibtext> </blist> <blist> <bibtext> Paterson, S. M., Laajala, T., & Lehtelä, P.-L. (2019). Counsellor students' conceptions of online counselling in Scotland and Finland. British Journal of Guidance & Counselling, 47 (3), 292 – 303. https://doi.org/10.1080/03069885.2017.1383357</bibtext> </blist> <blist> <bibtext> PDPC. (2020). Model artificial intelligence governance framework: Second edition (Personal Data Protection Commission of Singapore). https://<ulink href="http://www.pdpc.gov.sg/-/media/files/pdpc/pdf-files/resource-for-organisation/ai/sgmodelaigovframework2.pdf">www.pdpc.gov.sg/-/media/files/pdpc/pdf-files/resource-for-organisation/ai/sgmodelaigovframework2.pdf</ulink></bibtext> </blist> <blist> <bibtext> Pearcy, A. (2023, July 18). "It was as if my father were actually texting me": Grief in the age of AI. The Guardian.</bibtext> </blist> <blist> <bibtext> Peters, D., Vold, K., Robinson, D., & Calvo, R. A. (2020). Responsible AI – Two frameworks for ethical design practice. IEEE Transactions on Technology and Society, 1 (1), 34 – 47. https://doi.org/10.1109/TTS.2020.2974991</bibtext> </blist> <blist> <bibtext> Phillips, J. G., Currie, J., Ogeil, R. P., & Vaeau, F. (2020). Drug and alcohol counsellors' opinions of therapist negotiated location aware outpatient support using a mobile app. British Journal of Guidance & Counselling, 48 (2), 240 – 250. https://doi.org/10.1080/03069885.2020.1749981</bibtext> </blist> <blist> <bibtext> PWC China. (2023). Regulatory and legislation: China's interim measures on the management of generative artificial intelligence services officially implemented. https://<ulink href="http://www.pwccn.com/en/tmt/interim-measures-for-generative-ai-services-implemented-aug2023.pdf">www.pwccn.com/en/tmt/interim-measures-for-generative-ai-services-implemented-aug2023.pdf</ulink></bibtext> </blist> <blist> <bibtext> Reardon, S. (2023). AI chatbots could help provide therapy, but caution is needed. Scientific American.</bibtext> </blist> <blist> <bibtext> Russell, S. (2023, April 2). AI has much to offer humanity. It could also wreak terrible harm. It must be controlled. The Observer.</bibtext> </blist> <blist> <bibtext> Schick, N. (2023). Faking it – Navigating the new era of generative AI may be the most critical challenge to democracy yet. RSA Journal, 2023 (2), 40 – 43.</bibtext> </blist> <blist> <bibtext> Schuett, J., Reuel, A., & Carlier, A. (2023). How to design an AI ethics board. https://arxiv.org/pdf/2304.07249.pdf</bibtext> </blist> <blist> <bibtext> Sebri, V., Pizzoli, S. F. M., Savioni, L., & Triberti, S. (2020). Artificial intelligence in mental health: Professionals' attitudes towards AI as a psychotherapist. Annual Review of Cyber Therapy and Telemedicine, 18, 229 – 233.</bibtext> </blist> <blist> <bibtext> Shaji George, A., Hovan George, A. S., & Gabrio Martin, A. S. (2023). A review of ChatGPT AI's impact on several business sectors. Partners Universal International Innovation Journal, 1 (1), 9 – 23. https://<ulink href="http://www.researchgate.net/publication/368662952%5fA%5fReview%5fof%5fChatGPT%5fAI%27s%5fImpact%5fon%5fSeveral%5fBusiness%5fSectors">www.researchgate.net/publication/368662952%5fA%5fReview%5fof%5fChatGPT%5fAI%27s%5fImpact%5fon%5fSeveral%5fBusiness%5fSectors</ulink></bibtext> </blist> <blist> <bibtext> Shorey, S., Ang, E., Yap, J., Ng, E. D., Lau, S. T., & Chui, C. K. (2019). A virtual counseling application using artificial intelligence for communication skills training in nursing education: Development study. Journal of Medical Internet Research, 21 (10), e14658. https://doi.org/10.2196/14658</bibtext> </blist> <blist> <bibtext> Simpson, S. (2009). Psychotherapy via videoconferencing: A review. British Journal of Guidance & Counselling, 37 (3), 271 – 286. https://doi.org/10.1080/03069880902957007</bibtext> </blist> <blist> <bibtext> Smith, K., Moller, N., Cooper, M., Gabriel, L., Roddy, J., & Sheehy, R. (2022). Video counselling and psychotherapy: A critical commentary on the evidence base. Counselling and Psychotherapy Research, 22 (1), 92 – 97. https://doi.org/10.1002/capr.12436</bibtext> </blist> <blist> <bibtext> Stade, E. C., Stirman, S. W., Ungar, L. H., Boland, C. L., Schwartz, H. A., Yaden, D. B., Sedoc, J., DeRubeis, R. J., Willer, R., & Eichstaedt, J. C. (2023). Large language models could change the future of behavioural healthcare: A proposal for responsible development and evaluation. PsyArXiv. https://doi.org/10.31234/Osf.Io/Cuzvr</bibtext> </blist> <blist> <bibtext> Strickland, E., & Zorpette, G. (2023, June 21). The AI apocalypse: A scorecard. IEEE Spectrum. https://spectrum.ieee.org/artificial-general-intelligence</bibtext> </blist> <blist> <bibtext> Suleyman, M., & Bhaskar, M. (2023). The coming wave. Bodley Head.</bibtext> </blist> <blist> <bibtext> Thenral, M., & Annamalai, A. (2020). Telepsychiatry and the role of artificial intelligence in mental health in post-COVID-19 India: A scoping review on opportunities. Indian Journal of Psychological Medicine, 42 (5), 428 – 434. https://doi.org/10.1177/0253717620952160</bibtext> </blist> <blist> <bibtext> Timmons, A. C., Duong, J. B., Simo Fiallo, N., Lee, T., Vo, H. P. Q., Ahle, M. W., Comer, J. S., Brewer, L. C., Frazier, S. L., & Chaspari, T. (2022). A call to action on assessing and mitigating bias in artificial intelligence applications for mental health. Perspectives on Psychological Science, 174569162211344. https://doi.org/10.1177/17456916221134490</bibtext> </blist> <blist> <bibtext> Webb, T., Holyoak, K. J., & Lu, H. (2023). Emergent analogical reasoning in large language models. Nature Human Behaviour, 7 (9), 1526 – 1541. https://doi.org/10.1038/s41562-023-01659-w</bibtext> </blist> <blist> <bibtext> Weisel, K. K., Fuhrmann, L. M., Berking, M., Baumeister, H., Cuijpers, P., & Ebert, D. D. (2019). Standalone smartphone apps for mental health – A systematic review and meta-analysis. NPJ Digital Medicine, 2 (118), https://doi.org/10.1038/s41746-019-0188-8</bibtext> </blist> <blist> <bibtext> Wright, J. (2002). Online counselling: Learning from writing therapy. British Journal of Guidance & Counselling, 30 (3), 285 – 298. https://doi.org/10.1080/030698802100002326</bibtext> </blist> <blist> <bibtext> Zhou, S., Zhao, J., & Zhang, L. (2022). Application of artificial intelligence on psychological interventions and diagnosis: An overview. Frontiers in Psychiatry, 13. https://doi.org/10.3389/fpsyt.2022.811665</bibtext> </blist> </ref> <aug> <p>By Pei Boon Ooi and Graeme Wilkinson</p> <p>Reported by Author; Author</p> <p></p> <p>Dr. Ooi Pei Boon is a registered counsellor with the Malaysia Board of Counsellors and Associate Professor at Sunway University. As health behaviours change counsellor, her research interests are in the intersection of cyber and the well-being of individuals.</p> <p>Dr. Graeme Wilkinson is Tan Sri Jeffrey Cheah Distinguished Professor at Sunway University, Malaysia. His research interests are in the applications of artificial intelligence and the impact of new technologies on society. He is a fellow of the British Computer Society.</p> </aug> <nolink nlid="nl1" bibid="bib18" firstref="ref1"></nolink> <nolink nlid="nl2" bibid="bib68" firstref="ref2"></nolink> <nolink nlid="nl3" bibid="bib69" firstref="ref3"></nolink> <nolink nlid="nl4" bibid="bib31" firstref="ref4"></nolink> <nolink nlid="nl5" bibid="bib40" firstref="ref5"></nolink> <nolink nlid="nl6" bibid="bib48" firstref="ref6"></nolink> <nolink nlid="nl7" bibid="bib55" firstref="ref8"></nolink> <nolink nlid="nl8" bibid="bib10" firstref="ref9"></nolink> <nolink nlid="nl9" bibid="bib59" firstref="ref11"></nolink> <nolink nlid="nl10" bibid="bib76" firstref="ref12"></nolink> <nolink nlid="nl11" bibid="bib41" firstref="ref13"></nolink> <nolink nlid="nl12" bibid="bib21" firstref="ref14"></nolink> <nolink nlid="nl13" bibid="bib77" firstref="ref16"></nolink> <nolink nlid="nl14" bibid="bib75" firstref="ref17"></nolink> <nolink nlid="nl15" bibid="bib57" firstref="ref18"></nolink> <nolink nlid="nl16" bibid="bib50" firstref="ref19"></nolink> <nolink nlid="nl17" bibid="bib66" firstref="ref20"></nolink> <nolink nlid="nl18" bibid="bib23" firstref="ref22"></nolink> <nolink nlid="nl19" bibid="bib44" firstref="ref23"></nolink> <nolink nlid="nl20" bibid="bib19" firstref="ref24"></nolink> <nolink nlid="nl21" bibid="bib22" firstref="ref25"></nolink> <nolink nlid="nl22" bibid="bib30" firstref="ref26"></nolink> <nolink nlid="nl23" bibid="bib14" firstref="ref27"></nolink> <nolink nlid="nl24" bibid="bib20" firstref="ref28"></nolink> <nolink nlid="nl25" bibid="bib62" firstref="ref29"></nolink> <nolink nlid="nl26" bibid="bib39" firstref="ref30"></nolink> <nolink nlid="nl27" bibid="bib71" firstref="ref31"></nolink> <nolink nlid="nl28" bibid="bib17" firstref="ref32"></nolink> <nolink nlid="nl29" bibid="bib28" firstref="ref33"></nolink> <nolink nlid="nl30" bibid="bib72" firstref="ref34"></nolink> <nolink nlid="nl31" bibid="bib63" firstref="ref35"></nolink> <nolink nlid="nl32" bibid="bib53" firstref="ref36"></nolink> <nolink nlid="nl33" bibid="bib32" firstref="ref37"></nolink> <nolink nlid="nl34" bibid="bib11" firstref="ref39"></nolink> <nolink nlid="nl35" bibid="bib78" firstref="ref40"></nolink> <nolink nlid="nl36" bibid="bib33" firstref="ref41"></nolink> <nolink nlid="nl37" bibid="bib45" firstref="ref42"></nolink> <nolink nlid="nl38" bibid="bib26" firstref="ref43"></nolink> <nolink nlid="nl39" bibid="bib34" firstref="ref44"></nolink> <nolink nlid="nl40" bibid="bib43" firstref="ref45"></nolink> <nolink nlid="nl41" bibid="bib67" firstref="ref46"></nolink> <nolink nlid="nl42" bibid="bib25" firstref="ref47"></nolink> <nolink nlid="nl43" bibid="bib13" firstref="ref48"></nolink> <nolink nlid="nl44" bibid="bib65" firstref="ref49"></nolink> <nolink nlid="nl45" bibid="bib36" firstref="ref51"></nolink> <nolink nlid="nl46" bibid="bib47" firstref="ref52"></nolink> <nolink nlid="nl47" bibid="bib49" firstref="ref53"></nolink> <nolink nlid="nl48" bibid="bib35" firstref="ref54"></nolink> <nolink nlid="nl49" bibid="bib15" firstref="ref55"></nolink> <nolink nlid="nl50" bibid="bib74" firstref="ref56"></nolink> <nolink nlid="nl51" bibid="bib73" firstref="ref57"></nolink> <nolink nlid="nl52" bibid="bib61" firstref="ref58"></nolink> <nolink nlid="nl53" bibid="bib70" firstref="ref59"></nolink> <nolink nlid="nl54" bibid="bib38" firstref="ref60"></nolink> <nolink nlid="nl55" bibid="bib16" firstref="ref63"></nolink> <nolink nlid="nl56" bibid="bib42" firstref="ref64"></nolink> <nolink nlid="nl57" bibid="bib58" firstref="ref71"></nolink> <nolink nlid="nl58" bibid="bib46" firstref="ref72"></nolink> <nolink nlid="nl59" bibid="bib24" firstref="ref73"></nolink> <nolink nlid="nl60" bibid="bib56" firstref="ref74"></nolink> <nolink nlid="nl61" bibid="bib27" firstref="ref75"></nolink> <nolink nlid="nl62" bibid="bib52" firstref="ref76"></nolink> <nolink nlid="nl63" bibid="bib12" firstref="ref77"></nolink> <nolink nlid="nl64" bibid="bib60" firstref="ref78"></nolink> <nolink nlid="nl65" bibid="bib37" firstref="ref79"></nolink> <nolink nlid="nl66" bibid="bib54" firstref="ref80"></nolink> <nolink nlid="nl67" bibid="bib51" firstref="ref81"></nolink> <nolink nlid="nl68" bibid="bib29" firstref="ref82"></nolink> <nolink nlid="nl69" bibid="bib64" firstref="ref84"></nolink>
Header DbId: eric
DbLabel: ERIC
An: EJ1473691
AccessLevel: 3
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Enhancing Ethical Codes with Artificial Intelligence Governance -- A Growing Necessity for the Adoption of Generative AI in Counselling
– Name: Language
  Label: Language
  Group: Lang
  Data: English
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Pei+Boon+Ooi%22">Pei Boon Ooi</searchLink> (ORCID <externalLink term="https://orcid.org/0000-0002-7259-3381">0000-0002-7259-3381</externalLink>)<br /><searchLink fieldCode="AR" term="%22Graeme+Wilkinson%22">Graeme Wilkinson</searchLink> (ORCID <externalLink term="https://orcid.org/0000-0003-0709-222X">0000-0003-0709-222X</externalLink>)
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="SO" term="%22British+Journal+of+Guidance+%26+Counselling%22"><i>British Journal of Guidance & Counselling</i></searchLink>. 2025 53(1):66-80.
– Name: Avail
  Label: Availability
  Group: Avail
  Data: Routledge. Available from: Taylor & Francis, Ltd. 530 Walnut Street Suite 850, Philadelphia, PA 19106. Tel: 800-354-1420; Tel: 215-625-8900; Fax: 215-207-0050; Web site: http://www.tandf.co.uk/journals
– Name: PeerReviewed
  Label: Peer Reviewed
  Group: SrcInfo
  Data: Y
– Name: Pages
  Label: Page Count
  Group: Src
  Data: 15
– Name: DatePubCY
  Label: Publication Date
  Group: Date
  Data: 2025
– Name: TypeDocument
  Label: Document Type
  Group: TypDoc
  Data: Journal Articles<br />Reports - Descriptive
– Name: Subject
  Label: Descriptors
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Ethics%22">Ethics</searchLink><br /><searchLink fieldCode="DE" term="%22Artificial+Intelligence%22">Artificial Intelligence</searchLink><br /><searchLink fieldCode="DE" term="%22Governance%22">Governance</searchLink><br /><searchLink fieldCode="DE" term="%22Computer+Mediated+Communication%22">Computer Mediated Communication</searchLink><br /><searchLink fieldCode="DE" term="%22Counseling%22">Counseling</searchLink><br /><searchLink fieldCode="DE" term="%22Psychotherapy%22">Psychotherapy</searchLink><br /><searchLink fieldCode="DE" term="%22Foreign+Countries%22">Foreign Countries</searchLink><br /><searchLink fieldCode="DE" term="%22Decision+Making%22">Decision Making</searchLink><br /><searchLink fieldCode="DE" term="%22Algorithms%22">Algorithms</searchLink>
– Name: Subject
  Label: Geographic Terms
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22United+States%22">United States</searchLink><br /><searchLink fieldCode="DE" term="%22Australia%22">Australia</searchLink><br /><searchLink fieldCode="DE" term="%22United+Kingdom%22">United Kingdom</searchLink><br /><searchLink fieldCode="DE" term="%22Malaysia%22">Malaysia</searchLink>
– Name: DOI
  Label: DOI
  Group: ID
  Data: 10.1080/03069885.2024.2373180
– Name: ISSN
  Label: ISSN
  Group: ISSN
  Data: 0306-9885<br />1469-3534
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: The advent of generative Artificial Intelligence (AI) systems, such as large language model chatbots, is likely to have a significant impact in psychotherapy and counselling in the future. In this paper we consider the current state of AI in psychotherapy and counselling and the likely evolution of this field. We examine the ethical codes of practice for counselling in four countries in different parts of the world, namely the UK, the USA, Australia and Malaysia, and identify aspects of these codes that will need enhancement to reflect good AI governance. Using the Model Artificial Intelligence Governance Framework as an example, we have identified how the key elements of the AI framework relate to the core elements of the ethical codes, as a pointer to how such ethical codes will need to be enhanced if generative AI systems are to be adopted by the counselling profession.
– Name: AbstractInfo
  Label: Abstractor
  Group: Ab
  Data: As Provided
– Name: DateEntry
  Label: Entry Date
  Group: Date
  Data: 2025
– Name: AN
  Label: Accession Number
  Group: ID
  Data: EJ1473691
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=eric&AN=EJ1473691
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1080/03069885.2024.2373180
    Languages:
      – Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 15
        StartPage: 66
    Subjects:
      – SubjectFull: Ethics
        Type: general
      – SubjectFull: Artificial Intelligence
        Type: general
      – SubjectFull: Governance
        Type: general
      – SubjectFull: Computer Mediated Communication
        Type: general
      – SubjectFull: Counseling
        Type: general
      – SubjectFull: Psychotherapy
        Type: general
      – SubjectFull: Foreign Countries
        Type: general
      – SubjectFull: Decision Making
        Type: general
      – SubjectFull: Algorithms
        Type: general
      – SubjectFull: United States
        Type: general
      – SubjectFull: Australia
        Type: general
      – SubjectFull: United Kingdom
        Type: general
      – SubjectFull: Malaysia
        Type: general
    Titles:
      – TitleFull: Enhancing Ethical Codes with Artificial Intelligence Governance -- A Growing Necessity for the Adoption of Generative AI in Counselling
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Pei Boon Ooi
      – PersonEntity:
          Name:
            NameFull: Graeme Wilkinson
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 01
              Type: published
              Y: 2025
          Identifiers:
            – Type: issn-print
              Value: 0306-9885
            – Type: issn-electronic
              Value: 1469-3534
          Numbering:
            – Type: volume
              Value: 53
            – Type: issue
              Value: 1
          Titles:
            – TitleFull: British Journal of Guidance & Counselling
              Type: main
ResultId 1