Misleading AI claims becoming systemic business risk

Ethical AI washing occurs when organizations publicly promote commitments to fairness, transparency, accountability, or responsible AI governance without implementing meaningful internal mechanisms to support those claims. Codes of ethics, advisory boards, and glossy responsibility statements may exist, but they are often disconnected from operational decision-making, system design, or oversight processes. This practice is especially damaging because it exploits growing public concern about AI harms while offering little real protection.

A new study warns that many companies are overstating their AI capabilities, creating a growing gap between public claims and actual technology use.

The study, AI Washing and the Erosion of Digital Legitimacy: A Socio-Technical Perspective on Responsible Artificial Intelligence in Business, finds that exaggerated and symbolic AI claims are undermining trust in digital innovation. Instead of supporting responsible adoption, the research shows that AI washing is distorting competition, weakening accountability, and eroding confidence in what artificial intelligence can truly deliver.

How AI washing takes shape across modern businesses

AI washing, as defined by the author, is a set of practices through which organizations portray themselves as more AI-enabled, advanced, or ethically responsible than their underlying technologies justify. Unlike outright fraud, AI washing often operates in gray zones where claims are technically vague, difficult to audit, and strategically framed to signal innovation without offering verifiable detail.

The author identifies several drivers behind this phenomenon. One is information asymmetry. AI systems are inherently opaque to most stakeholders, including customers, investors, regulators, and even internal decision-makers. This opacity allows firms to rely on broad labels such as AI-powered, machine learning-based, or intelligent systems without clearly specifying the scope, autonomy, or impact of the technology involved.

Another driver is institutional pressure. As AI becomes a marker of competitiveness, firms face growing expectations to demonstrate AI adoption regardless of their actual readiness. In sectors ranging from finance and healthcare to retail and logistics, signaling AI capability is increasingly tied to market valuation, access to capital, and reputational standing. In this environment, symbolic compliance can appear more attractive than costly technical investment.

The study proposes a four-part typology of AI washing practices. The first involves marketing and branding exaggeration, where basic automation or rule-based software is presented as advanced AI. The second focuses on technical capability inflation, in which firms overstate the sophistication, autonomy, or learning capacity of their systems. The third relates to strategic signaling aimed at investors and partners, using AI narratives to influence funding, mergers, or partnerships. The fourth, and most concerning according to the study, is ethical AI washing.

Ethical AI washing occurs when organizations publicly promote commitments to fairness, transparency, accountability, or responsible AI governance without implementing meaningful internal mechanisms to support those claims. Codes of ethics, advisory boards, and glossy responsibility statements may exist, but they are often disconnected from operational decision-making, system design, or oversight processes. This practice is especially damaging because it exploits growing public concern about AI harms while offering little real protection.

These forms of AI washing are rarely isolated. In many cases, they reinforce one another, creating a comprehensive legitimacy strategy that blends technical ambiguity, ethical signaling, and innovation rhetoric. Over time, this strategy reshapes how stakeholders interpret AI claims, lowering expectations for evidence and normalizing symbolic compliance.

Digital legitimacy at risk as hype replaces accountability

The author frames digital legitimacy as the perceived alignment between an organization’s technological claims, its actual practices, and broader societal expectations. AI washing disrupts this alignment by allowing performative narratives to substitute for verifiable substance.

At the organizational level, AI washing can deliver short-term benefits. Firms may attract investment, command higher valuations, or gain competitive visibility by positioning themselves as AI leaders. However, the study finds that these gains are fragile. As discrepancies between claims and reality emerge, companies face reputational damage, loss of stakeholder trust, and potential legal exposure. Internally, inflated narratives can also create misalignment between leadership expectations and technical capacity, leading to poor strategic decisions.

At the industry level, AI washing distorts competition. Firms that invest heavily in genuine AI development must compete with rivals who rely on symbolic signaling at a fraction of the cost. This dynamic discourages long-term innovation and rewards rhetorical skill over technical competence. Over time, it can slow meaningful progress by reducing incentives for rigorous development, testing, and governance.

At the system level, the consequences are even more severe. AI washing fuels hype cycles that inflate expectations beyond what current technology can deliver. When reality fails to match these promises, public trust erodes, and regulatory backlash becomes more likely. The study warns that widespread AI washing could contribute to a legitimacy crisis, where stakeholders become skeptical of AI claims altogether, including those made by genuinely responsible actors.

The research also highlights the particular vulnerability of ethical AI discourse. As regulators, civil society, and consumers demand responsible AI, firms increasingly use ethics language as a reputational shield. Without clear standards or enforcement, ethical commitments become a low-cost signaling tool rather than a driver of behavioral change. This undermines the very idea of responsible AI by turning it into a branding exercise.

AI systems are not just technical artifacts but are embedded in organizational structures, governance regimes, and cultural narratives. Legitimacy is produced through interaction between technology, communication, and institutional norms. When these elements drift out of alignment, trust becomes performative rather than grounded.

Why regulation and governance lag behind AI claims

One reason AI washing persists, the study argues, is the gap between technological development and governance capacity. Existing regulatory frameworks struggle to keep pace with rapid innovation and often focus on outcomes rather than claims. As a result, misleading AI narratives can flourish even when systems fall short of their implied capabilities.

The research notes that current approaches to AI governance emphasize principles such as transparency and accountability but rarely specify how claims should be substantiated. This creates a paradox. Firms are encouraged to communicate about AI responsibly, yet there are few consequences for vague or exaggerated statements. The absence of shared definitions for terms like AI-powered or intelligent system further complicates enforcement.

The author calls for clearer taxonomies and measurable criteria to distinguish between different levels of AI capability. Without such distinctions, stakeholders cannot meaningfully assess claims, and regulators lack the tools to intervene. The study suggests that explainability, documentation, and third-party audits could play a role, but only if they are tied to enforceable standards rather than voluntary disclosure.

The paper also highlights the role of media and consulting ecosystems in amplifying AI washing. Reports, rankings, and trend analyses often rely on self-reported data or surface-level indicators, reinforcing symbolic compliance. In this environment, narratives of AI leadership can spread faster than technical scrutiny, further weakening accountability.

Importantly, the study does not argue against AI adoption or innovation. Instead, it warns that unchecked AI washing threatens the conditions under which responsible innovation can thrive. When legitimacy is decoupled from substance, trust becomes unstable, and the social license to deploy AI erodes.