The Pitfalls of AI Washing: How Exaggerated AI Claims Can Harm Businesses and Mislead Consumers

The release of OpenAI’s ChatGPT in 2022 sparked a surge of interest in generative AI, leading many tech companies and startups to tout AI-driven innovations. However, the reality often falls short of these claims. According to a study, the percentage of startups mentioning AI in pitches increased from 10% in 2022 to over 25% in 2023. Despite this, only 4.4% of American businesses were actually using AI, as found by a US Census Bureau survey in November last year.

This phenomenon, termed ‘AI washing,’ mirrors ‘greenwashing’ where companies exaggerate eco-friendliness. AI washing involves businesses overstating their use of AI or marketing less advanced technology as sophisticated AI. Misleading advertisements about AI capabilities or non-functional features are also considered AI washing.

One notable example includes Google’s Gemini, which was presented as an advanced AI chatbot capable of real-time image recognition. However, it was later revealed that the demo video was not real-time but stitched together using text prompts. Amazon also faced scrutiny when its ‘Just Walk Out’ technology, claimed to use AI for cashier-less checkouts, was found to rely on human reviews.

Even big brands like McDonald’s and Coca-Cola have been implicated. McDonald’s withdrew its AI-driven drive-thru technology after numerous order errors, while Coca-Cola’s AI-generated flavor failed to impress consumers. Additionally, several apps claiming AI capabilities were merely repackaging OpenAI’s technology.

The trend of AI washing can mislead businesses seeking genuine AI solutions, hinder innovation, and pose data security risks to consumers. To combat this, the US Federal Trade Commission (FTC) advises companies to honestly assess their AI claims. Simply using AI in development does not equate to having an AI-powered product. Transparency and authenticity are crucial in avoiding the pitfalls of AI washing.