Artificial intelligence: Tech marvel or ‘morally problematic’? | Analysis

“From the very beginning, AI has been a marketing campaign, and a very successful one. Artificial intelligence was a sexy term, mainly used to get funding,’ proclaimed Hubert Entienne, CEO of Quintessence AI and founding chair of the Organisation Committee of the Paris Conference on AI and digital ethics, during his keynote at Campaign360.

Entienne, an AI ethicist formerly at the helm of Meta’s Global Generative AI Ethics division, now pioneers the field of ‘computational philosophy’. His innovative approach founded with Facebook’s AI research, merges deep philosophical inquiry with advanced computational techniques to analyse and interpret online social behaviours.

As AI has evolved, so has its narrative, shared Entienne. What began in the 1950s as a nascent field driven by large-scale computer research and sabermetrics, has morphed into a multifaceted ecosystem spanning myriad applications. From the Turing machine’s early data encryption to modern algorithms predicting social trends, AI’s journey has been marked by significant milestones that reflect both advances and ethical quandaries.

Entienne delved deep into this journey, emphasising the dual nature of AI’s evolution. “As modern AI technologies have become more generalised and multimodal, integrating various data types into cohesive systems, AI’s applicability and accessibility have broadened, transforming it from a niche field into a mainstream technology,” he explained. Yet, he was quick to remind the audience of AI’s roots in strategic marketing, which have played a significant role in its widespread adoption.

However, the notion that AI is merely a technological triumph does not hold without scrutiny, Entienne insisted, bringing to light the darker aspects of AI, particularly the biases embedded within. “AI systems are trained on massive amounts of data, and embedded in that data are societal biases,” he pointed out, highlighting the risk of perpetuating discrimination as well as the urgent need to address bias in AI models and hold organisations accountable.

Illustrating the practical applications and ethical dilemmas of AI, Entienne discussed how AI could be used responsibly, like in collecting sensitive data to ensure algorithms are fair across different communities. Conversely, he critiqued applications designed to predict personal attributes from images, which he deemed “morally problematic”.

The challenge of transparency and accountability in AI systems also occupied a significant part of his address. “Operating within a ‘black box,’ these systems offer limited interpretability, making it difficult to understand how decisions are made and who is responsible for them,” Entienne remarked. He emphasised the development of ‘explainable AI’ which could ensure fairness, accuracy, and accountability, particularly in critical domains like healthcare and autonomous vehicles.

Entienne also explored the emerging issues of AI-generated content, creativity, and ownership. “As AI advances, questions arise regarding the ownership rights of AI-generated art and potential copyright infringement,” he stated, pointing to the broader implications for cultural sensitivities and the representation of various races, sexualities, and genders in AI-generated content.

Moreover, the ethical implications of AI extend to social manipulation and misinformation, particularly through technologies like deepfakes. “These technologies can spread misinformation, manipulate public opinion, and amplify social divisions, necessitating vigilance and countermeasures to address these challenges effectively,” he warned.

As he summarised his insights, Entienne noted that the marketing industry is at a pivotal point in AI development; as systems become increasingly sophisticated, the need for transparency is more crucial than ever. He urged practitioners to commit to ethical usage by understanding, researching, reasoning, and deciding their uses for AI, and encouraging collaboration across industries to share best practices and develop guiding regulations. When applied correctly, he stressed, AI can become a powerful compass in navigating the complex marketing landscape.

Entienne concluded by quoting Welsh poet Dylan Thomas’ infamous poem ‘Do not go gentle into the good night’, urging marketers to resist complacency and not to passively accept the advancements and implications of AI as they are, but to continually strive for improvement and accountability in how such technologies can be developed and implemented.