The Rise of 'AI Washing' and Its Implications
'AI washing' refers to companies making exaggerated claims about their use of AI. Amazon's 'Just Walk Out' technology highlighted the issue of AI washing. Some companies falsely claim to use AI or overstate its effectiveness.
This phenomenon, akin to "green washing" in the environmental context, raises concerns about the accuracy and transparency of AI implementation. Amazon's "Just Walk Out" technology, which allows customers to pick things in actual grocery shops and be immediately invoiced, was one high-profile case that raised awareness of this issue. However, investigations indicate that human workers in India manually reviewed a major number of the transactions, contradicting Amazon's assertions of a totally AI-powered system.
While the specifics of the Amazon case may differ, it shows a bigger issue of firms overstating their AI capabilities. Some organisations erroneously claim to use AI while actually using less complex computational approaches. Others exaggerate the usefulness of their AI solutions in comparison to existing methodologies, or depict AI chatbots as fully functional when they are not. This false behaviour not only affects consumer trust, but also makes it difficult for investors to find truly creative companies.
The prevalence of AI washing has been rising. According to OpenOcean, a UK and Finland-based investment company, the number of digital start-ups mentioning AI in their pitches climbed from 10% in 2022 to more than a quarter by 2023. This proportion is likely to rise above one-third this year. The urge to get finance and appear cutting-edge has prompted some founders to overestimate their AI capabilities, resulting in a substantial gap between claims and genuine AI-driven outcomes.
The lack of a broadly accepted definition of AI exacerbates the issue. Different conceptions of AI make it difficult to regulate and hold firms accountable for deceptive statements. Douglas Dick, UK head of emerging technology risk at KPMG, highlights the ambiguity of AI, arguing that the word is used broadly and loosely with no clear point of reference. This ambiguity allows AI washing to thrive, perhaps resulting in firms overpaying for technology and services or failing to meet operational goals.
Regulators are beginning to take note of AI washing. The US Securities and Exchange Commission (SEC) sued two financial advice firms with making false and misleading representations concerning their use of artificial intelligence (AI). The SEC's stern attitude indicates that individuals who breach AI standards can expect to face additional fines and sanctions. AI washing is already addressed in the UK by guidelines and laws like as the Advertising Standards Authority's code of conduct, which prohibits marketing messages that significantly mislead consumers.
While AI offers some advantages, experts warn against naively deploying it in all sectors. Sandra Wachter, an Oxford University technology and regulation professor, doubts the necessity of employing AI for every work and emphasises AI's often-overlooked environmental impact. She advocates for a more sophisticated approach, taking into account specific tasks and industries where AI can be truly advantageous.
AI washing may become less of an issue in the long term as AI gets more prevalent. According to Advika Jalan, head of research at MMC Ventures, the branding strategy of being AI-powered would most certainly lose its differentiation aspect as AI becomes more mainstream, similar to declaring "we're on the internet."
'AI washing' refers to companies making exaggerated claims about their use of AI.
Amazon's 'Just Walk Out' technology highlighted the issue of AI washing.
Some companies falsely claim to use AI or overstate its effectiveness.
Source: BBC