AI Ethics

Trend Overview

Impact:

Significant

Adoption:

< 5 Years

Focus Areas:

Health, Safety & Future of Work Physical & Digital Security

Trend Clusters:

Artificial Intelligence Digital Backbone

Sectors:

The trend of Artificial Intelligence Ethics (AI Ethics) refers to the increasing focus on the ethical considerations and implications of AI technologies and ensuring their responsible development and deployment. Within this trend we also explore legislative developments of AI platforms, and the implications of data safety and algorithm security.

AI ethics is an emerging field concerned with ensuring the responsible development, deployment, and use of AI technologies. Its development accelerated in response to growing public awareness of AI's ethical implications, fueled by high-profile incidents, debates around democratizing it, and developments in regulatory frameworks to address issues such as bias, privacy, explainability, and transparency in AI systems.

Topics to be addressed include explainability, trustworthiness, transparency, robustness, and interpretability. The trend plays a critical role in the logistics industry, especially around these topics, as the number of AI applications increases within logistics and the supply chain. In particular, data privacy concerns must be addressed and ethical use of AI platforms must be ensured.

As the topic of AI ethics will impact most industries, this in turn will have significant impact on the logistics industry.

With increasing reliance on AI-powered systems for managing logistics operations, there’s a growing need to address concerns regarding the privacy and security of sensitive data. AI ethics standards will emphasize the importance of robust data protection measures to safeguard customer information and proprietary data.

Company compliance regulations for employees have existed for decades. These must, however, be amended to also address compliant use of company data when AI is integrated with collaboration tools and with company portals. Here at DHL, we recognize company compliance regulations must align with national laws. This opens a new domain in the world of compliance and legal clauses/contractual obligations for employees.

Source: Based on DataCamp (2023): Explainable AI - Understanding and Trusting Machine Learning Models

Relevance to the Future of Logistics

Trustworthy & Explainable AI

Explainable AI is the ability to trace back an algorithm to the data that it is built on, and find the logical chain of association from secure and trustworthy data, and its traceable training, to build a deep learning algorithm. In this logic similar to a mathematical equation, we are able to determine and explain with high confidence the output of what AI has either generated or analyzed. Explainable AI can offer insights into the factors influencing demand forecasts, enabling better decision making and accountability. The explainability of those outputs is found in the processing of consumer data which, if tampered with, renders the output untrustworthy. Here we see an emphasis on the importance of data security and protection against cyberattack and hacking.

AI can be made explainable and trustworthy through understanding:

Intention: AI systems are constructed by humans to make decisions based on historical data or real-time information. Predetermined responses are embedded in the AI systems.

Intelligence: The ability to make intelligent decisions with AI systems is facilitated by combining machine learning and data analytics. AI isn’t intelligent like a human being. Human intelligence is closest to what a machine can approximate.

Adaptivity: AI systems compile information to make decisions based on collecting and adapting to new information. AI systems can improve the outcome of decision making with information they learn from real-time data.

Next use case

Ethical End User / Use

Consumer ethical behavior around the use of AI systems includes compliant and ethical content creation and not publishing deep fakes of any persons. Examples of this are the creation of deep fake videos of public figures posted with false messages about companies, governments, or other public entities. Such occurrences would involve hacking and accessing data which is “stolen” and not intended for use beyond approved groups of people, and using this data to generate untrustworthy outputs.

In logistics, employees have access to customer data, trade secrets, and other sensitive company information which can be used to build algorithms. Non-compliant behavior could stem from this. There is a growing need to implement robust measures to secure data processing and accessibility.

This can also be said for use cases such as AI-generated route optimization, or questioning outputs that appear biased or discriminatory. End users may report instances of bias to the relevant authorities or organizations encouraging the development of fairer algorithms and mitigating potential harm to marginalized groups. Some consumers actively engage in co-design processes or participate in feedback mechanisms to contribute to the development of more ethical AI systems. By sharing their perspectives, concerns, and values, all consumers can help shape AI technologies that align with ethical principles and better service diverse needs and interests.

Previous use case Next use case

AI Legislation & Democratization

General Data Protection Regulation (GDPR), applicable in the European Union (EU) and the European Economic Area (EEA), regulates the processing of data and imposes strict requirements on data controllers and processors. Logistics providers using AI systems must comply with GDPR principles, ensuring transparency, lawfulness, and fairness in the processing of all data.

The European Commission has published ethical guidelines for trustworthy AI, emphasizing principles such as fairness, transparency, accountability, and societal benefit.

While not legally binding, these guidelines influence the development and deployment of AI systems in various sectors, including logistics. Other legislative implementations, such as the California Consumer Privacy Act (CCPA), underline the economic impact of AI and the need for regulatory measures to ensure ethical use. It is essential to keep pace. For example, in Singapore, government funding of $20 million Singapore dollars (US$15.3 million) now enables people aged above 40 to learn AI models.

Democratization of AI involves making AI technologies accessible, affordable, and easy to use for the widest range of individuals and organizations, irrespective of their technical expertise or financial resources. Key aspects of AI democratization include providing accessible tools and platforms, offering affordable pricing models, providing education and training opportunities, fostering community collaboration, and emphasizing ethical considerations in AI development and deployment.

Previous use case

Challenges

The logistics industry faces challenges related to algorithmic bias, which can lead to unfair treatment, inefficiencies, and discrimination in supply chain operations (sanctions and geopolitical events fuel public algorithms).

Managing sensitive data within supply chains raises ethical dilemmas regarding data privacy, security, and ownership, particularly when AI systems are involved in data processing and analysis.

Ensuring accountability and transparency in AI-driven decision-making processes within logistics requires clear mechanisms for understanding, auditing, and explaining the reasoning behind AI-generated recommendations and actions.

Managing sensitive data within supply chains raises ethical dilemmas regarding data privacy, security, and ownership, particularly when AI systems are involved in data processing and analysis.

Challenge 1 Challenge 2 Challenge 3

Outlook

Over the next five years, the logistics and supply chain industries are likely to face heightened regulatory scrutiny regarding AI ethics, with governments and industry bodies introducing new laws, guidelines, and standards to address ethical concerns related to data privacy, algorithmic bias, and accountability in AI-driven logistics operations.

Companies operating in logistics and the supply chain are expected to increasingly adopt and implement ethical AI frameworks and best practices to mitigate risks, build trust with stakeholders, and ensure responsible AI deployment. This includes integrating principles such as fairness, transparency, accountability, and societal benefit into AI development, deployment, and use.

As awareness of AI ethics grows, there will be a greater emphasis on ethical considerations throughout the supply chain, including sourcing, manufacturing, distribution, and customer engagement. Companies will leverage AI technologies to enhance supply chain visibility, traceability, and sustainability while addressing ethical concerns such as human rights violations, environmental impact, and employee welfare.

This trend should be CAREFULLY monitored,with implementations available for many use cases today.