Salesforce, UL & DNV: How to Use AI Ethically & Responsibly

Artificial intelligence (AI) is creating new opportunities across sectors, from supporting healthcare diagnoses to expanding access to personalised education and helping safeguard cultural diversity.

Yet the rapid development and deployment of AI also raises questions around bias, privacy, safety, human rights and environmental impact.

In November 2021, UNESCO’s 193 Member States adopted the first global standard on AI ethics, establishing principles including transparency, fairness, environmental sustainability and human oversight.

As AI becomes embedded in products, workplaces and critical infrastructure, ethical and responsible use increasingly depends on how systems are designed, tested, governed and monitored throughout their lifecycle.

From Turing to generative AI

The foundations of modern AI can be traced to the 1950s, when English mathematician and computer science pioneer Alan Turing explored whether machines could demonstrate intelligent behaviour.

AI subsequently developed through several stages, including machine learning in the 1980s, when systems began using historical data to improve their performance.

A major milestone came in 1997 when IBM’s Deep Blue defeated world chess champion Garry Kasparov, demonstrating the potential of machines to perform complex tasks.

During the 2010s, deep learning expanded AI’s capabilities by using increasingly sophisticated neural networks, while today’s generative AI models can create text, images, code and other content. 

This evolution has also expanded the ethical questions surrounding AI, making issues such as bias, transparency, data privacy, accountability and human oversight increasingly important.

“Artificial intelligence and generative AI may be the most important technology of any lifetime,” says Marc Benioff , chair, CEO and Co-founder, Salesforce

UL Solutions puts AI safety under scrutiny

UL Solutions is addressing responsible AI through safety evaluation and certification for AI-enabled products.

Its UL 3115 Outline of Investigation provides a framework for assessing AI-based products before and during deployment, covering areas including technical performance, ethics and governance.

The evaluation considers risks such as malfunction, misuse, malicious interference and data compromise, while its ethical pillar addresses fairness, bias and data privacy.

UL Solutions also assesses transparency, explainability, oversight and lifecycle management, helping organisations evaluate how AI behaves in real-world applications.

Ethical AI must’s:

• AWS: Humans must retain final decision-making power and responsibility, often practiced through a “human-in-the-loop” approach.

• UNESCO: User data must be securely safeguarded and respected throughout the entire lifecycle of the AI model.

• SAP: Users and operators need clear insight into how an algorithm reaches its conclusions, avoiding opaque “black-box” decisions.

• UNESCO: Technologies must be robust against attacks and engineered to prevent physical, psychological or societal harm.

• IBM: Systems must treat all individuals and groups equitably, actively working to prevent discrimination or bias stemming from training data or algorithms.

Its AI algorithm reproducibility programme further examines whether AI technologies can deliver predictable and repeatable outcomes, while marketing claim verification can assess whether claims made about AI-enabled products are supported by evidence.

DNV focuses on AI throughout its lifecycle

DNV’s approach to trustworthy and responsible AI centres on assurance throughout the technology’s lifecycle rather than compliance alone.

Its recommended practice for AI-enabled systems provides guidance for understanding whether AI performs as intended and how legal and ethical requirements can be translated into practical assurance processes.

DNV also emphasises the importance of identifying stakeholders affected by AI, including people who may never directly interact with a system but experience its outcomes.

In healthcare, for example, doctors may use AI to support diagnoses while patients experience the consequences of those decisions.

In maritime operations, passengers, operators, developers and regulators can similarly have different relationships with AI systems.

DNV’s approach therefore considers data, sensors, algorithms, digital twins, human oversight and ongoing monitoring as interconnected elements of responsible AI.

Salesforce builds trust into AI development

Salesforce takes a product-focused approach to trusted AI, embedding responsible AI practices across design, development, testing and governance.

Its Office of Ethical and Humane Use brings together responsible AI, ethical use policy and accessibility teams, with AI systems undergoing intake, triage, review, testing and implementation before deployment. 

“There’s no question we are in an AI and data revolution, which means that we’re in a customer revolution and a business revolution,” says Clara Shih , previous CEO for Salesforce AI, current Senior Advisor and Former Head of Business AI at Meta.

“But it’s not as simple as taking all of your data and training a model with it.

“There’s data security, there’s access permissions, there’s sharing models that we have to honour.

“These are important concepts, new risks, new challenges and new concerns that we have to figure out together.” 

Testing includes adversarial testing, content safety assessments, accessibility checks, employee trust testing and large-scale stress testing to identify potential failure modes.

Salesforce also uses platform-level guardrails covering areas such as privacy, accuracy, safety, autonomy and monitoring, alongside controls designed to detect prompt injection and keep AI agents within their intended scope.

Transparency measures include AI disclosure, model cards and evaluations focused on safety, privacy, truthfulness and fairness, while human-AI handoffs and accessibility are incorporated into the product experience.

DNV’s partners

United Nations Global Compact

DNV became a signatory to the United Nations Global Compact in 2003, committing to embed its ten principles covering human rights, labour, environment and anti-corruption across its strategy and operations. DNV also aligns its annual reporting with the framework and collaborates with the initiative on opportunities and solutions supporting the UN Sustainable Development Goals.

World Business Council for Sustainable Development

DNV has been an active member of the World Business Council for Sustainable Development since 1999, with its Group President and CEO serving on the Executive Committee. Through the partnership, DNV advocates for sustainable and responsible business practices while participating in climate and energy programmes focused on developing and deploying technologies and solutions.

Red Cross

DNV has partnered with the Red Cross since 2004 to support humanitarian capacity and provide volunteering opportunities for employees. The partnership includes annual financial contributions to the Norwegian, Netherlands and British Red Cross organisations, alongside additional donations. DNV employees also contribute technical, digital, safety and sustainability expertise to relevant Red Cross projects.