Responsible AI in the Consumer Goods and Retail Industry

Download PDF version

• CG&R companies who used automated pricing should be prepared to handle concerns over whether such algorithms may lead to an anticompetitive environment, as or potential bias due to flawed datasets.
• CG&R companies must be wary of relying purely on algorithms to set prices. This may lead to price-surging during times of high demand, but can also lead to a loss of trust and goodwill among consumers if it results in pricing some out of the market entirely.

• CG&R companies should pay attention to data used to train their AI systems to avoid creating new entry points for hackers to attack, as well as ensuring that any sensitive personal data contained within datasets is subject to additional levels of protection.
• Required security of training AI data is twofold — companies must be able to ensure not only that the data cannot be stolen or accessed by unauthorized parties, but also that it cannot be added to by unauthorized parties to ensure the continued integrity of the AI’s decision-making.

• Transparency and explainability are key elements for ensuring that consumers trust an AI system’s insights, understand when they are engaging with an AI, and can challenge the outcomes and decisions of the AI system. This can help brands decrease the risk of error and misuse of AI.
• CG&R companies making use of AI that interacts with consumers should disclose when AI is being used and be prepared to explain any automated and algorithmic decisions and any associated data driving those decisions.

This rapidly evolving area has a great deal of international legislative development in progress. Companies should ensure internal policies and guidelines comply with the latest regulations and guidelines.