How Can We Ensure Responsible AI in Marketing?

Organizations are increasingly investing in AI systems and tools to enhance business processes and maximize value. AI’s integration with businesses is expanding globally, with recent estimates suggesting that about 42% of companies are using AI in some way, and AI investment is forecasted to reach $200 billion globally by 2025. Generative AI has accelerated this further, with creative industries in particular being transformed by its wide-ranging applications.

Much of the attention so far has been paid to applications that pose a significant risk to human well-being, particularly given the number of high-profile harms associated with the use of AI without the appropriate safeguards in recent years. However, seemingly low-risk applications of AI can still have important business implications if they go wrong. This is particularly true for industries like marketing, where the use of AI tools is becoming ubiquitous, and is directly linked to a firm’s performance and market reach.

In this blog post, we provide an overview of the ways in which AI tools are used in marketing, the associated risks and harms, and how responsible AI can be operationalized within organizations using AI in their marketing.

Key takeaways:

• While AI in marketing is typically not considered high-risk to society, its trust and efficacy can have important business implications
• Uses of AI in marketing include SEO, sentiment analysis, predictive analytics, and personalization
• Highlighting the need for responsible AI in marketing, a number of fines have been issued over the use of AI-driven marketing tools. Notably, Luxembourg’s Data Protection authority fined Amazon a whopping €746 million over its ad targeting tool
• Responsible AI in marketing can be operationalized through external third-party auditing,  a focus on compliance, and creating an AI inventory. There also needs to be commitment from C-suite executives, with Chief Marketing Officers playing a pivotal role

How is AI being used in marketing?

AI is being used in marketing in a variety of ways – such as in analyzing customer behavior, automating tasks, personalizing content, and predicting outcomes. Using AI tools in marketing can lead to many benefits for businesses, from faster decision-making and accurate KPI measurement to a better ROI on marketing investments and enhanced customer experience. Below are several examples of the ways in which AI tools are commonly used in marketing.

Synthetic Content Generation

From music to news to programming, there is now a plethora of content generator tools that can save marketers time and money. These tools can be used to generate material like blogs, video captions, marketing messages, emails, and other forms of collateral.

Audience Segmentation and Sentiment Analysis

AI tools are often used to detect behavioral and other patterns among customers to allow marketing campaigns to be more targeted and personalized. AI tools can be used to segment audiences into categories according to behaviors, interests, traits, and preferences. Achieving this granularity in customer insights drives audience engagement and purchases, ultimately leading to greater ROIs.

A key benefit of AI-driven sentiment analysis is that it further allows companies to draw comprehensive generalizations from large volumes of textual data so that they can better understand how different groups of people perceive them, enabling them to analyze vast volumes of data and identify patterns that may not be intuitive to humans. From this, insights can be translated into business value by using key themes in customer feedback to inform business strategy, marketing campaigns, and product features.

Search Engine Optimisation (SEO)

AI tools can significantly enhance SEO by helping marketers create and optimize content that leads to higher rankings. In particular, SEO tools can be used to enhance keyword searching, content writing, idea generation, and video creation, all of which are otherwise time-consuming tasks for organizations to perform.

AI Chatbots for Customer Relationship Management (CRM)

AI technologies are increasingly being used to enhance sales and customer experience. In particular, chatbots are being trained to respond to customer requests more effectively, leading to increased customer satisfaction. Companies’ integration of such tools could lead to a significant reduction in costs and a simultaneous growth in revenue – for instance, e-commerce sales through AI chatbots are expected to reach $142 billion by 2024, with a Gartner study reporting that chatbots will be the main customer service channel for almost 25% of organizations by 2027.

Predictive Analytics and Dynamic Pricing

An Aberdeen study found that companies that identify customer needs through predictive analytics can increase revenue by 21% each year, versus an average of 12% without the use of predictive analytics. Predictive analytics uses algorithms to forecast possible future buys using data from customers’ past purchases, including date, time, and season of purchase.

Email marketing campaigns can also leverage these tools to deliver personalized messages to customers, which can lead to a six times higher transaction rate than campaigns without personalized messaging. AI tools can also be used to optimize real-time pricing by using data gathered from customer behavior, competitors, and current company inventory to increase profitability and help ensure demand is being covered.

Programmatic Advertising

Programmatic advertising, which is the automation of ad purchasing and placements, leads to higher conversion rates by allowing marketers to deliver more relevant ads based on customer history and preferences. While traditional ad purchasing can be labor-intensive and time-consuming, the use of AI tools in programmatic advertising can lead to significant ROIs for big and small enterprises alike as this is a more efficient, targeted, and scalable method. Unsurprisingly, organizations invest significant resources into programmatic advertising -  Global spending on programmatic advertising reached almost $558 billion in 2023, and is expected to cross $700 billion by 2026.

Why is it important to have responsible AI in marketing?

Deploying AI tools without robust safeguards in place can cause companies significant financial, legal, and reputational harm. For example, an error in Google’s Bard chatbot resulted in a loss of 100 billion dollars in market capitalization (9% stock drop) for the parent company in early 2023. Companies can also incur financial costs in other ways – in 2014, Uber faced harsh criticism after the app raised prices due to increased demand during a café siege in Sydney. To respond, the company had to offer free rides to try to win back consumer trust and its reputation, showing the multiple risks that companies face from misusing AI systems.

Penalties for AI marketing tools

As well as financial impacts from damage to trust, there have also been several penalties issued to  organizations for the misuse of AI marketing tools.

For example, the Information Commissioner’s Office (ICO), the supervisory authority for data protection in the UK,  fined Scottish company CRDNN Ltd a record £500,000 in 2020 for a breach of Regulations 19 and 24 of the Privacy and Electronic Communications Regulations (PECR), which is the maximum amount available for a breach of the PECR. The fines came as CRDNN made over 19 million unsolicited automated direct marketing calls in a span of just four months and refused to facilitate requests from those who had opted out of future calls, showing that there was no proper consent.

More recently, the ICO imposed issued a £20,000 fine to Royal Mail for its an automated marketing tool in 2022. The tool, Eloqua, was sending emails to customers who opted out of receiving them, and this was found to be a violation of Regulation 22 of the PECR, which requires valid consent for marketing materials.

More over, in 2021, the Luxembourg Data Protection Authority CNPD (Commission Nationale pour la Protection des Données) fined Amazon a whopping €746 million for General Data Protection Regulation (GDPR) non-compliance. Amazon appealed this fine in January 2024, which the court is currently hearing. According to the CNPD, the targeted ad system Amazon employed was found to be processing personal data and conducting behavioral advertising without proper consent, but under local laws, details of the case have not yet been made publicly available as the appeal process is underway. This penalty marks the second-largest fine imposed under the GDPR since it was enacted, the largest being a historic €1.2 billion fine for Meta for violating privacy rules by transferring the personal data of people in the EU to the US.

Similar to Luxembourg’s case against Amazon, the French Data Protection Authority CNIL (Commission nationale de l'informatique et des libertés) fined Google €50 million in 2019 under the GDPR for the lack of a valid legal basis in the processing of personal data, especially with regards to ad personalization. Specifically, Google was found to have breached transparency and information obligations under Articles 12 and 13, where the CNIL found Google’s claim that it rightfully obtains user consent for data processing for ad personalization to be invalid, and also observed that these violations are continuous breaches of the GDPR rather than a one-time infringement.

How can responsible AI be operationalized in marketing?

It is crucial for companies using AI tools in marketing to ensure that they are operationalizing responsible AI practices to protect themselves from reputational, financial, and legal damage and business performance – through the use of better tools, companies can expand their reach, increase customer loyalty, and adapt to evolving trends more quickly.

In this section, we outline some of the best practices that organizations can implement to operationalize responsible AI in marketing.

External AI governance mechanisms

Third-party algorithm audits and assurance methods are becoming increasingly popular and required under various laws and can strengthen organizations’ AI governance. Algorithm Auditing is the research and practice of assessing, mitigating, and assuring an algorithm’s safety, legality, and ethics. Key verticals to consider are:

• Bias – assessment of whether outputs favor one group through the use of gendered language for example, or contain harmful stereotypes. Ensuring that outputs are fair and non-toxic can avoid reputational damage and strengthen business performance by ensuring that messaging is appropriate and targeted to the right audiences.
• Explainability – whether the outputs of AI marketing tools can be explained from the inputs, and whether users who are analyzed by algorithms are informed of such. As well as being conducive to greater accountability – which can help to increase trust –  explainability also acts as a bulwark for regulatory compliance, allowing enterprises to create robust documentation to maximize regulatory compliance and streamline decision-making.‍
• Data governance (privacy) – whether data minimization and data stewardship principles have been implemented, particularly when processing client or otherwise sensitive or personal data. By having governance mechanisms that track where data is coming from, its quality, and how it is being used and manifesting on a model level, the threat of reputational damage and legal liability is lowered.
• ‍Robustness – how well the model is able to withstand malicious attacks or spam reviews and poor data. By ensuring that AI systems are robust against attempts at data poisoning or attempts to sabotage the model’s effectiveness, marketers can ensure that they maximize the value of their tools and gain actionable insights that are not swayed by outliers or poor-quality data.
• ‍Efficacy – an assessment of how well an AI model performs relative to its use case. A model that is slow, resource-intensive, or that doesn’t perform well leads to poor ROI. By measuring the efficacy of a model, and taking steps to improve it if needed, marketers can ensure their AI is an asset, rather than a liability.

Compliance with relevant regulation

The regulatory landscape surrounding AI is fast expanding across the globe and is targeting a variety of applications of AI. Companies are expected to do their due diligence and comply or risk reputational hits along with hefty fines, as could be seen from the examples above. Keeping up-to-date with relevant regulations and taking steps to comply is an essential component of responsible AI.

Moreover, taking voluntary action to implement AI governance frameworks can give organizations a competitive advantage and increase trust in their use of AI, both internally and externally.

Create an AI inventory

With the vast applications of AI in marketing, marketing departments are likely to be using a range of AI models across different activities. Creating a real-time inventory of the AI systems being used where can help to avoid efforts and resources being duplicated, streamline risk management by being able to develop a clear road map, and maximize compliance by easily being able to determine the tools in the scope of different laws.

C-suite engagement

Responsible AI is only possible with the commitment of top executives within a company, and establishing these initiatives is increasingly becoming a board-level and C-suite priority. From the top, there must be active efforts to deploy responsible AI best practices and foster cross-functional collaboration. The onus lies on more than just developers and engineers – harms can be predicted and mitigated by pushing teams across various disciplines to collaborate, such as social scientists with data scientists.

Chief Marketing Officers play a vital role in this - the push for collaboration from the top is critical to ensure there are internal efforts towards identifying and mitigating key risks. minimalization utilize.

Achieve responsible AI with Holistic AI

AI’s transformative power in marketing and driving companies' bottom-line is clear. However, without due diligence, deploying AI tools can come at large costs as implementing AI can be costly, complex, and risky. As investments to integrate AI grow, it is important to ensure that companies are maximizing the value they can reap from this technology.

Schedule a demo of Holistic AI’s Governance Platform for a 360° evaluation of your company’s AI systems to ensure responsible development and deployment.