AI washing (the practice of overstating or misrepresenting the use of AI technologies) is a growing issue that can mislead consumers and investors, and poses significant risks. Regulators in Australia, the US, and the EU are taking steps to address these practices, signalling the need for companies to ensure transparency and integrity in marketing their AI capabilities.
In 2016, Amazon launched its ‘Just Walk Out’ checkout technology, which enabled customers to leave a store without needing to checkout as the purchase was automatically charged. Amazon stated that the technology used computer vision, sensor fusion and deep learning to track all purchases made by customers. However, reports emerged earlier this year alleging that the technology was more human than artificial, as it relied on more than 1,000 workers in India to manually review around 70% of transactions in 2022. Amazon has since responded to say that the workers review and improve the system, not run it.
Regardless of the specific details of the Amazon case, it highlights a growing trend known as ‘AI washing’. While some companies genuinely adopt AI technologies to improve processes, others merely attach labels such as ‘AI-powered’ without meaningful use of the technology, which raises significant ethical and legal considerations, potentially impacting compliance, transparency, and trust.
What is AI washing?
AI washing refers to the practice of companies overstating or misrepresenting their use of artificial intelligence technologies in their business, products or services. Similar to greenwashing, where businesses make false or misleading environmental claims, AI washing has the potential to mislead or deceive customers and investors, by exploiting the growing interest in AI and presenting the company as more innovative and technologically advanced than it is.
In fact, some studies suggest that startups using AI are between 15-50% more likely to gain investment than those that don’t mention the use of AI.1 Common ways that companies engage in AI washing include:
- advertising certain products as being AI driven or AI powered without specifying how AI is applied or if it improves performance meaningfully
- claiming automation through AI, when operations are heavily reliant on human involvement
- adding non-essential AI features to otherwise traditional products to position them as advanced or cutting-edge
- promising performance improvements that are not realistic given the level of technology involved.
Risks with AI washing
Companies that engage in AI washing could in time face regulatory scrutiny, legal penalties, and reputational damage. Some prevalent risks include:
- breach of consumer law – misleading marketing claims about AI functionality are likely to violate consumer protection laws. For instance, these claims might breach the Australian Consumer Law (ACL), which prohibits misleading or deceptive conduct, and certain false and misleading representations. Breaches of consumer protection laws can lead to substantial fines, with penalties in Australia reaching up to the greater of $50,000,000, three times the benefit reasonably attributable to the contravention, or if the benefit cannot be determined, 30% of adjusted turnover during the relevant period
- breach of contract – if a company promises AI-based outcomes that are not delivered, it faces the risk of breach of contract which may result in costly litigation or loss of market reputation. This could result in costly litigation, with contract counterparties seeking damages or specific performance. Beyond financial costs, a breach of contract over AI capabilities could lead to a loss of customer or partner trust, damaging long-term relationships and potentially impacting the company’s standing in its industry
- investor risks – startups or other companies that overstate their AI capabilities may attract investors or partnerships based on false premises, which could result in regulatory action. This risk is heightened as investors and regulators increasingly scrutinise AI-related claims, given the sector’s growth and evolving regulatory landscape. Companies found to have misled investors could face regulatory action, such as fines or sanctions from securities regulators, in addition to the reputational damage of losing investor trust.
Enforcement in Australia, the US and the EU
While there has been no official statement from the Australian Competition and Consumer Commission (ACCC) regarding AI Washing, global regulators are beginning to respond to AI washing through both policy and direct enforcement (most of which is currently in the early stages).
The US
In the U.S., the Federal Trade Commission (FTC) launched “Operation AI Comply” in September 2024, an enforcement initiative targeting AI washing and other deceptive practices related to AI. The FTC has issued several guidance documents, warning companies, among other things, “don’t say you use AI tools if you don’t.” The agency has also described AI as a “hot marketing term” that advertisers are prone to “overuse and abuse.” Additionally, the U.S. Securities and Exchange Commission (SEC) has taken enforcement action against several companies over misleading AI claims, including settlements with two investment advisers for allegedly making false statements about their use of AI.
The EU
The EU AI Act, which came into effect on 1 August 12024, does not specifically address AI washing (for more information on the EU AI Act, read our article linked here). However, it introduces stringent transparency requirements, compelling organisations to disclose their use of AI more clearly. While not AI specific, the EU’s unfair commercial practices directive (UCPD) prohibits misleading practices and statements, including statements which contains false or untrue information or are likely to deceive the average consumer. Claims about AI which are false or are likely to deceive are likely to fall foul of the UCPD.
Australia
In Australia, the ACCC has previously demonstrated its readiness to act against misleading or false claims, for example by naming greenwashing as a compliance and enforcement priority for 2022-2023. The ACCC’s Digital Platforms Services Inquiry has also stressed the importance of tackling deceptive practices in the digital space, particularly those involving emerging technologies like AI.
More recently, the Treasury released a discussion paper evaluating whether the ACL remains adequate to protect consumers from the risks posed by AI. While the paper doesn’t focus specifically on AI washing, it signals that we can expect new consumer protections related to AI and emerging technologies in the future.
Key takeaways
AI washing is a growing issue. While legitimate AI adoption can provide real benefits, misleading claims not only erode trust but also expose businesses to legal, financial, and reputational risks. Although specific guidance from the ACCC has not yet been released, it’s likely that a company’s statements about the use of AI will be subject to the same consumer protection standards as other goods and services, with misleading claims about AI capabilities facing scrutiny under the Australian Consumer Law.
Ultimately, as regulations evolve, companies must prioritise transparency and integrity in marketing AI capabilities to ensure sustainable innovation and avoid the consequences of misleading stakeholders.
For more information, please contact Alistair Newton and Isabel Roach.
1Bernard Marr, Forbes Article, Spotting AI Washing: How Companies Overhype Artificial Intelligence (Forbes Article, April 2024).
