In recent years, artificial intelligence (AI) has surged in capability and adoption, from generative AI chatbots to decision-making systems. This rapid growth brings immense opportunities and new ethical challenges. Stakeholders – from regulators to consumers – are increasingly concerned about AI’s social impacts. Notably, government leaders have declared that companies developing AI “have a responsibility to ensure their products are safe”, urging industry to uphold the highest standards so that innovation “doesn’t come at the expense of … rights and safety”. For businesses, this imperative goes beyond compliance; it aligns with corporate social responsibility (CSR) obligations to society.
One useful framework to address these challenges is ISO 26000, the international standard for social responsibility. Although published in 2010 (before today’s AI boom), ISO 26000 provides a comprehensive CSR lens that can guide ethical AI governance. This article examines how organizations can integrate AI governance with CSR principles—particularly the seven core subjects of ISO 26000—to ensure ethical AI practices. We’ll explore real-world examples (2024 onwards), case studies, and current best practices that demonstrate experience, expertise, authoritativeness, and trustworthiness (E-E-A-T) in this critical intersection of technology and responsibility.
What is ISO 26000? ISO 26000 is an international standard that offers guidance on social responsibility for all types of organizations. Unlike some ISO standards, it’s not certifiable, but it outlines key principles and seven core subjects of CSR that serve as a roadmap for ethical business conduct. These seven core areas are: organizational governance, human rights, labor practices, the environment, fair operating practices, consumer issues, and community involvement and development. In essence, ISO 26000 encourages companies to be accountable for their impacts on society and the environment in these domains.
Applying ISO 26000 to AI: The relevance to AI becomes clear when we consider that AI’s deployment can have far-reaching effects across all these areas. Researchers have noted that while AI is transforming industries and society, it also “faces ethical, moral, privacy, and security issues”, making it urgent to apply a social responsibility management system to AI. By examining AI through ISO 26000’s seven lenses, organizations can systematically identify risks and address them. In fact, scholars have analyzed the risks of AI in each of ISO 26000’s seven aspects and propose corresponding countermeasures. This structured approach ensures that as AI technologies evolve, they do so in a way that aligns with societal values and CSR principles.
Below, we break down each of the seven core CSR subjects and discuss how they connect to AI governance and ethical AI, with real-world examples and practices from 2024 that illustrate these concepts in action.
Effective organizational governance is the foundation for responsible AI. In the context of ISO 26000, organizational governance refers to the system of oversight, policies, and processes by which an organization makes decisions and upholds accountability. Applying this to AI means establishing clear leadership and oversight for AI initiatives. Companies should integrate AI ethics into their corporate governance structures – for example, by creating AI ethics committees, appointing a Chief AI Ethics Officer, or extending board oversight to include AI risks.
Real-world practice shows this is gaining traction. Many leading tech companies have developed internal AI ethics boards or advisory councils. For instance, IBM formed an AI Ethics Board and has imbued its development processes with ethical checkpoints. IBM’s leaders note that “some of our leading clients are bringing [sustainability] requirements into their enterprise AI governance frameworks”, indicating that businesses are beginning to link AI performance with broader CSR goals. Another example is the voluntary commitments made by seven AI giants (including Google, Microsoft, and OpenAI) in July 2023 to implement safety, security, and transparency testing before releasing AI products. These moves underscore how governance mechanisms—whether internal policies or industry pledges—can ensure AI is developed responsibly and with oversight.
Good AI governance under ISO 26000 also means transparency and accountability in AI decision-making. Organizations should openly communicate their AI ethics principles and be transparent about AI system capabilities and limitations. Accountability mechanisms (such as audit trails for AI decisions and public reporting on AI impact) help build trust. A lack of governance can lead to public backlash or regulatory action. Conversely, companies that proactively govern AI (through ethics guidelines, risk assessments, and stakeholder engagement) demonstrate expertise and trustworthiness, aligning with E-E-A-T by taking responsibility for AI’s outcomes.
Respect for human rights is a cornerstone of CSR and is directly challenged by AI in areas like bias, privacy, and surveillance. ISO 26000 emphasizes that organizations should respect all human rights – from equality and non-discrimination to privacy and freedom of expression. AI systems, if unchecked, can inadvertently infringe on these rights. For example, AI algorithms used in hiring, lending, or law enforcement have shown tendencies to perpetuate bias, leading to discrimination against protected groups. In 2023, the U.S. Equal Employment Opportunity Commission (EEOC) settled its first AI discrimination lawsuit, involving a recruitment AI that automatically rejected older job applicants, violating age discrimination laws. This case (EEOC v. iTutorGroup) highlights how AI can pose risks to the right to equal opportunity, and it reinforces that companies must vet AI tools for fairness.
Privacy is another human right at stake. AI systems often rely on massive personal data, raising concerns about consent and data protection. A notable example occurred in March 2023, when Italy’s data protection authority temporarily banned ChatGPT over privacy violations. The agency found an “absence of legal basis” for OpenAI’s mass data collection and cited inadequate age protections. OpenAI had to respond with remedies before service was restored. Similarly, the upcoming EU AI Act (entering into force August 2024) is rooted in safeguarding fundamental rights: it will ban AI uses that pose an “unacceptable risk” to human rights (such as social scoring or invasive surveillance) and require strict oversight for high-risk AI systems. These developments show that aligning AI practices with human rights is not just ethical – it’s increasingly a legal expectation.
To uphold human rights in AI, companies should conduct due diligence on their AI systems, akin to human rights impact assessments. This means evaluating algorithms for bias, ensuring AI decisions can be explained (to uphold due process rights), and protecting user privacy through robust data governance. Tech companies like Microsoft have adopted comprehensive human rights policies that extend to AI and supply chains. By proactively addressing bias and privacy, organizations demonstrate expertise in AI ethics and build public trust. Ethical AI design (e.g. inclusive training data, fairness audits, privacy-by-design) is thus a direct application of CSR principles to modern technology.
The labor practices core subject of ISO 26000 covers fair and decent work conditions, employee well-being, and social protections. AI is transforming the workplace, bringing both improvements and challenges to labor practices. On one hand, AI can enhance worker safety (e.g. predictive maintenance reducing accidents) and take over repetitive tasks, potentially freeing employees for higher-value roles. On the other hand, AI-driven automation threatens job displacement and can affect morale if not managed responsibly. CSR demands that companies address these impacts ethically.
A headline example came in 2023 when IBM’s CEO announced a pause in hiring for roles that could be automated by AI, noting that roughly 7,800 jobs might be replaced by AI in the coming years. This statement, while forward-looking, exemplified workers’ fears that AI could lead to layoffs without proper transition planning. Responsible use of AI in labor practices means companies should mitigate the social impact of automation. This includes retraining and upskilling programs to help employees adapt to new AI-augmented roles, transparent communication about workforce changes, and careful deployment of AI in a way that complements rather than purely replaces human labor. For instance, some organizations have pledged not to resort to abrupt AI-based layoffs and instead focus on reskilling employees for new roles that AI creates – a strategy aligned with CSR values of fairness and care for employees.
Another labor aspect is workplace surveillance and employee rights. AI-powered monitoring (from warehouse worker tracking to AI analyzing employee emails) can intrude on privacy or increase stress. CSR-aligned governance would set boundaries on such practices, ensuring they are transparent, necessary, and respect dignity at work. In 2023, labor unions even negotiated guardrails on AI usage. The Hollywood writers’ and actors’ strikes, for example, resulted in new contract terms that limit the use of generative AI for scripts and digital likeness of actors – ensuring humans retain creative control and receive compensation when AI is used. These cases underscore the importance of involving employees (and their representatives) in decisions about AI deployment.
In summary, aligning AI with fair labor practices means viewing employees as key stakeholders in AI adoption. By supporting workers through the AI transition – providing training, maintaining fair labor standards, and engaging workers in policy development – companies demonstrate experience in managing technological change responsibly. This approach builds trust internally and externally, showing that AI innovation can go hand-in-hand with social responsibility to employees.
Environmental stewardship is a well-established pillar of CSR, and it now extends to the footprint of AI systems. The environment core subject in ISO 26000 calls for responsibility in areas like resource use, pollution, and climate change mitigation. AI might not have a smokestack, but training and running AI models consume significant energy and resources. Data centers powering AI contribute to carbon emissions, and the hardware supply chain has environmental impacts. Thus, “green AI” is becoming part of AI governance – aiming to reduce energy use and use cleaner resources.
Consider the scale of AI’s footprint: Training a single large AI model can emit hundreds of tons of CO₂. A recent study reported that training GPT-3 (a 175-billion-parameter model) consumed about 1,287 MWh of electricity and produced 502 metric tons of carbon emissions – roughly equivalent to driving 112 gasoline cars for a year. Moreover, once deployed, popular AI services can draw even more power in aggregate. In 2024, Google revealed its data center emissions jumped 48% since 2019, partly due to the integration of AI into its products. Each user query to an AI model like ChatGPT can use 10 times the electricity of a standard Google search. These statistics highlight why companies must factor environmental costs into AI strategies.
CSR-aligned AI governance involves taking steps to minimize AI’s environmental impact. Practical measures include: using energy-efficient algorithms and hardware, running AI workloads in regions with renewable energy, optimizing code to require less computation, and offsetting carbon emissions from AI projects. Tech companies have started to act – for example, IBM reported in 2023 that 74% of the electricity powering its data centers is from renewable sources. IBM also developed smaller, efficient AI models (its “Granite” models at 13B parameters) to achieve tasks with less energy. Similarly, Google and Microsoft are investing heavily in renewable energy for cloud computing and improving data center cooling and efficiency.
Beyond reducing negatives, AI can also be leveraged for environmental benefits as part of CSR. AI is used to optimize energy usage in buildings, improve supply chain efficiency, and model climate change solutions. For instance, AI helps companies track and reduce their carbon footprint – IBM’s Envizi platform uses AI to help firms with ESG (environmental, social, governance) reporting and energy management. By aligning AI innovation with sustainability goals (as experience shows in the examples above), companies fulfill their CSR duty to the planet. This alignment enhances their authority and trustworthiness in the eyes of stakeholders who increasingly demand climate action. Ultimately, treating environmental responsibility as integral to AI governance ensures that “smart” technologies are also sustainable technologies.
“Fair operating practices” in ISO 26000 refer to the ethics of an organization’s conduct – including anti-corruption, fair competition, and respect for law in business dealings. When deploying AI, companies must ensure that these ethical business standards are upheld. AI systems can introduce new compliance risks or even be misused for unethical purposes, so robust controls are needed.
One area of concern is fair competition and antitrust. AI algorithms (especially in pricing and market analysis) could inadvertently enable collusion or unfair market dominance. For example, if multiple companies rely on the same smart pricing algorithm, there’s a risk they might all raise prices in unison (knowingly or not), harming consumers and violating competition laws. Regulators have taken note: in early 2024, U.S. lawmakers introduced the Preventing Algorithmic Collusion Act to explicitly bar companies from using algorithms to collude on prices. Around the same time, the Federal Trade Commission warned that using AI tools for pricing doesn’t exempt businesses from antitrust laws, cautioning that even an agreement to share “pricing algorithms can still be unlawful” if it leads to price-fixing. These actions signal that authorities will hold companies accountable for AI-driven anti-competitive behavior. CSR-minded firms, therefore, should proactively audit their AI for compliance with fair competition rules and implement guidelines to prevent such issues.
Another aspect is anti-corruption and integrity. AI could potentially be used to detect fraud and corruption (e.g. through anomaly detection in transactions), which is a positive application aligned with CSR. However, AI could also be misused to facilitate unethical behavior – for instance, generating deepfake documents or automating bribery through complex supply chain systems. Organizations should extend their codes of ethics to cover AI usage, ensuring that AI systems are not aiding any unethical practices. This might include restrictions on AI-generated content use (to prevent fraud or misinformation), strict oversight of AI decisions in sensitive areas like finance, and transparency with regulators.
Compliance and governance frameworks are evolving to include AI. Many companies are now incorporating AI governance into their overall GRC (governance, risk, compliance) programs, performing risk assessments on AI deployments similar to other operational risks. By doing so, they show expertise in navigating new legal/ethical terrain. A culture of ethics from the top can guide data scientists and engineers to consider legal and social implications of their AI solutions. In practice, this could mean mandatory ethics training for AI developers and checklists to ensure AI projects undergo legal review. Companies like Patagonia, known for ethical business, exemplify how a strong ethical culture influences decision-making (Patagonia integrates social/environmental criteria in business decisions). Adapting such culture to the AI era means upholding fairness and integrity in every AI-augmented business process.
In summary, fair operating practices in AI governance ensure that innovation doesn’t outpace ethics. By actively preventing anti-competitive behavior, respecting intellectual property, combating fraud, and complying with all relevant laws, businesses align their AI activities with CSR values of honesty and fairness. This commitment to ethical conduct enhances the company’s authority and trust – regulators, partners, and the public can see that the organization will do the right thing even as technology changes.
The consumer issues core subject of ISO 26000 covers consumer rights, safety, truthful communication, and data protection in the context of an organization’s products or services. AI systems often directly interface with consumers – think of AI-driven products like chatbots, recommendation algorithms, autonomous vehicles, or even medical AI tools. Ensuring these AI-enabled offerings are safe, transparent, and respect consumer rights is a critical part of CSR in the AI age.
One major concern is product safety and quality. AI can introduce new kinds of risks. For example, flaws in an AI system might lead a medical device to give a wrong diagnosis or cause a self-driving car to misinterpret an obstacle – potentially harming consumers. Companies have a responsibility to thoroughly test AI products and disclose their limitations. The voluntary AI commitments made by leading tech firms in 2023 explicitly included pledges for internal and external security testing of AI models before release. This reflects a CSR approach of “safety first” for consumers. Similarly, under the EU AI Act, certain AI systems (like chatbots or deepfake generators) will require transparency labels so users know they are interacting with AI. Being honest and clear with consumers aligns with ISO 26000’s guidance on truthful marketing and information.
Privacy remains a paramount consumer issue. As illustrated by the ChatGPT incident in Italy, consumers (and regulators) object when their personal data is used without consent. Compliance with data protection regulations (GDPR, etc.) is the minimum; leading companies go further by giving users control, such as data opt-outs or explainable AI features that let users understand how their data influences outcomes. In 2024, we see browsers and apps offering AI-powered features with on-device processing options, highlighting growing sensitivity to privacy.
Another consumer issue is misinformation and fairness in AI outputs. Generative AI can produce incorrect or even defamatory content. A high-profile case occurred in June 2023 when an AI chatbot (ChatGPT) fabricated accusations against a radio host, leading him to sue OpenAI for defamation. While the legal question of AI’s liability is still evolving, the lesson for companies is clear: they must put safeguards to prevent and correct harmful outputs. This can include content moderation, user feedback loops, and limitations on AI responses in high-stakes domains. Providing clear disclaimers about AI limitations (as OpenAI does, noting the chatbot “may occasionally generate incorrect information”) is also a responsible practice, though not sufficient on its own.
From a CSR perspective, treating consumers fairly in the age of AI means prioritizing user well-being and rights at every stage. This can be as simple as ensuring AI customer service bots hand off to a human when queries are complex or emotional, or as involved as offering remedies if an AI-driven decision (like a credit denial) was found to be faulty. Companies like Microsoft and Google have published AI responsible use principles that include commitments to fairness, transparency, and accessibility for users. Such public commitments, when backed by action, demonstrate trustworthiness. They help consumers feel confident that the organization’s use of AI is not exploitative but aimed at genuine value and respect for the customer.
The final core subject of ISO 26000 is community involvement and development – essentially, how an organization contributes to and engages with the communities in which it operates. In the realm of AI, this translates to involving stakeholders (beyond just customers and employees) in AI initiatives and ensuring that AI benefits society at large, not only the company’s bottom line.
One way businesses address this is through stakeholder engagement on AI ethics. Forward-thinking companies consult external experts, civil society, and impacted communities when developing AI that could significantly affect the public. For example, some tech firms have set up external AI advisory councils to get input from ethicists, human rights advocates, and community representatives. This practice aligns with CSR by respecting the interests of all stakeholders. It’s also reflected in the multi-stakeholder approach of initiatives like the Partnership on AI, a nonprofit coalition (founded by tech companies, academics, and NGOs) that produces guidelines on AI fairness and societal impacts. By participating in such coalitions or public forums, companies show accountability and openness to societal feedback on AI projects.
Investing in AI for social good is another facet of community development. Organizations can leverage their AI expertise to support projects that address social or environmental challenges, often in partnership with nonprofits or governments. A shining example is IBM’s Sustainability Accelerator program. Launched in 2021, this pro bono initiative applies IBM’s AI technology and expertise to help vulnerable communities tackle environmental threats. In 2024, IBM issued new grants under this program focused on developing “resilient cities”, aiming to use AI to bolster urban communities against climate and infrastructure challenges. Such efforts not only directly benefit communities but also help companies build experience in applying AI ethically in real-world contexts. They demonstrate the company’s expertise and experience (the first “E” in E-E-A-T) in solving tangible problems, which in turn bolsters its reputation.
Community involvement also means addressing the societal discourse on AI. In 2023 and beyond, AI has raised broad public questions about job futures, ethical boundaries, and the impact on local communities (like whether AI might widen economic inequalities or enable harmful surveillance). A responsible organization engages in this discourse honestly. That could involve publishing research on AI impacts, sponsoring educational programs on AI literacy, or working with local institutions to ensure AI deployments (like smart city projects) are done with community consent and benefit. For instance, some tech companies are partnering with universities and NGOs to improve AI literacy and access in underrepresented communities, ensuring the benefits of AI are widely shared.
Ultimately, integrating community involvement in AI governance reflects a shift from a purely technocratic approach to a human-centric approach. By listening to community concerns and contributing to societal well-being, companies build trust at the grassroots level. They show that AI innovation is not happening in a bubble but is guided by empathy and social purpose. This trust can be invaluable, especially when controversies arise; an organization known for its community engagement is more likely to be given the benefit of the doubt and to navigate challenges successfully.
The convergence of AI, corporate social responsibility, and governance is not just a theoretical ideal – it’s becoming a practical necessity for organizations aiming to thrive in the modern era. By viewing ethical AI through the lens of ISO 26000’s CSR principles, companies gain a well-rounded perspective that covers internal governance structures, core human values, workforce implications, environmental sustainability, ethical business conduct, consumer protection, and community well-being. As we’ve seen with the examples from 2023–2024, neglecting any of these aspects can lead to regulatory backlash, reputational damage, or even legal liability. Conversely, embracing them can enhance a company’s reputation, foster innovation, and build public trust.
Critically, aligning AI with CSR is an ongoing journey. Technology evolves rapidly – consider that issues like generative AI misinformation or algorithmic collusion were minor concerns a few years ago, but are front and center now. Therefore, governance frameworks must be adaptive, guided by enduring ethical principles. Frameworks like ISO 26000 provide stability in values (e.g. respect for rights, accountability, transparency) even as specific best practices change. They encourage organizations to institutionalize E-E-A-T qualities: Experience in applying ethics to real projects, Expertise in both AI and social issues, Authoritativeness by adhering to global standards and regulations, and Trustworthiness through consistent ethical conduct and engagement.
In practical terms, businesses should incorporate ISO 26000-based checks into their AI project lifecycle. For example, during AI design, ask: Are we respecting human rights and fairness? During deployment: Have we considered consumer privacy and safety? For strategy: How does AI adoption align with our sustainability commitments and stakeholder expectations? By internalizing these questions, CSR considerations become part of day-to-day AI governance.
The year 2025 and beyond will likely bring even more emphasis on responsible AI. Regulations such as the EU AI Act will start to enforce risk management and transparency obligations, and stakeholders will expect demonstrable action, not just promises. Those organizations that have proactively bridged AI and CSR will be well-positioned – they will not only comply with new rules but also differentiate themselves as ethical leaders in their industry.
In conclusion, the ethical concerns of AI can indeed be managed in harmony with CSR principles. By using ISO 26000 as a guiding framework, companies can ensure that their pursuit of AI innovation aligns with societal values and sustainable development goals. This alignment is key to unlocking AI’s benefits while upholding the trust of employees, customers, regulators, and communities. Responsible AI governance is not a one-time checklist but a corporate culture of “doing well by doing good” – one that treats ethical AI as integral to business success and social responsibility in the 2024–2025 landscape and beyond.
A1: ISO 26000 is a global CSR standard outlining seven principles of social responsibility. It helps organisations govern AI ethically by addressing impacts on human rights, labour, sustainability, and fairness.
A2: AI systems affect decision-making in areas like hiring, surveillance, customer service, and resource consumption—directly impacting stakeholders and requiring ethical oversight aligned with CSR values.
A3: In 2023, Italy banned ChatGPT temporarily for violating privacy regulations. This shows that insufficient data governance in AI can result in legal bans, reputational damage, and loss of trust.
A4: Through ethical audits, stakeholder engagement, sustainability-aligned data infrastructure, AI ethics training, human oversight systems, and public commitments to responsible innovation.
A5: ISO 26000 is voluntary, but aligning AI development with it improves compliance readiness, public trust, and alignment with frameworks like the EU AI Act, ESG reporting, and UN SDGs.
