Throughout company profits calls, investor displays and advertising pitches, “artificial intelligence” has change into the buzzword of selection. But a troubling trend lies below the hype. Many claims hugely overstate precise AI sophistication, deceptive folks about true functions, long term results and doable harms.
A working example is the new 600% proportion value surge of Allbirds, after the once-trendy sustainable shoes industry issued a obscure announcement in April 2026 that it might pivot to AI. Within the coming months, the corporate plans to rename itself NewBird AI and surrender its standing as a public get advantages company.
As a student who research company sustainability, I see parallels between this “AI washing” phenomenon – when corporations oversell some great benefits of AI whilst glossing over the dangers – and the greenwashing pattern within the fresh previous, when corporations claimed to decide to sustainability however didn’t enact basic alternate. Common deception used to be rampant, with companies spending way more on inexperienced advertising than on precise sustainability enhancements. And the ones efforts steadily backfired on each the firms and the communities they served. Much more worrisome: AI washing’s fast upward thrust and common adoption will most probably eclipse the greenwashing tendencies.
How we were given right here
AI washing is prospering as a result of corporations and policymakers forget about 4 essential ideas. Those shortfalls, previously, additionally characterised greenwashing.
First, AI tips lack standardization. By way of 2019, 84 units of AI ethics ideas and tips had already been revealed. By way of 2023, this quantity had exploded to greater than 200 – a multitude of voluntary frameworks from corporations, analysis establishments and public organizations.
Making issues worse is that the U.S. recently is dependent upon fragmented AI laws, with maximum being voluntary. The Trump management has in most cases sided with Giant Tech to ward off efforts at state or federal legislation. At a world stage, some of the few exceptions is the Eu Union AI Act, in all probability some of the complete frameworks, however its implementation gained’t be absolutely phased in till 2027 or later.
Within the early 2000s, company sustainability confronted a equivalent credibility disaster. Each and every corporate measured sustainability another way, making comparisons inconceivable and greenwashing simple. The step forward got here handiest when standardized, industry-specific metrics allowed significant benchmarking. Projects such because the International Reporting Initiative and Sustainability Accounting Requirements Board established commonplace metrics for measuring environmental affect, social accountability and high quality of governance, identified by way of the shorthand ESG.
The U.N. local weather alternate summits, like this one in Brazil in 2025, have introduced a world discussion board for policymakers and industry leaders on local weather and sustainability problems.
AP Picture/Fernando Llano
When corporations should document carbon emissions the use of the similar technique, as an example, or reveal exertions prerequisites the use of equivalent classes, buyers can examine efficiency, establish laggards and allocate capital accordingly. This push made comparisons imaginable and deception tougher, despite the fact that it nonetheless wasn’t foolproof. As an example, a 2023 United Countries Surroundings Programme document at the fast-fashion {industry} discovered that many corporations proceed to make “vague and inflated” sustainability claims.
2d, there aren’t any complete frameworks within the U.S. that require companies to pass judgement on how AI impacts them in a subject matter method and publicly reveal the ones affects. Examples of AI-driven subject matter affects come with whether or not algorithmic bias shapes industry results, or whether or not selections on find out how to use AI techniques elevate importance for shareholders and the general public.
As a substitute, AI governance stays ruled by way of the slender interior circle of businesses that construct the AI techniques, whilst affected communities hardly have a say in figuring out which AI affects are subject matter sufficient to warrant public consideration. As an example, Giant Tech corporations like Google, Microsoft, Apple, NVIDIA and others adhere to their very own AI governance tips, with slightly little public enter.
The advance of sustainability ideas gives some examples of find out how to construct those frameworks. The EU’s Company Sustainability Reporting Directive calls for over 50,000 corporations to officially evaluation which sustainability subjects are subject matter to their stakeholders, after which reveal that knowledge. Those efforts attempt to make sure that duty is obvious throughout whole provide chains.
Whilst nowhere just about as complete, U.S. laws such because the 2010 Dodd-Frank monetary reform and California’s legislation requiring reporting on statewide greenhouse gasoline emissions supply a equivalent blueprint that U.S. policymakers may just construct on if they selected.
A 3rd downside is the overall loss of third-party verification, making AI washing trivially simple. Efficient disclosure approach reporting all subject matter affects – no longer simply cherry-picked successes.
In apply, AI audits can range dramatically in rigor, scope and technique. One auditor may behavior intensive trying out throughout demographic teams, analyze decision-making and validate the standard of coaching records. Every other may merely assessment documentation and settle for corporate explanations at face price. Given the number of AI auditing fashions in the market, other auditors would possibly use incompatible methodologies, making effects inconceivable to check. If corporations followed third-party accreditation techniques to evaluate how they use AI, they’d assist ensure that the duty that self-reported claims can’t fit.
In contrast, there used to be affordable development on this appreciate as corporations followed ESG ideas. As an example, establishments such because the Carbon Disclosure Challenge and International Reporting Initiative have a community of companions that provide unbiased verification. Those suppliers, qualified below world requirements, check company sustainability records towards rigorous standards. That method, they give you the assurance that we could corporations display the development had to release sustainable finance and mitigate felony dangers. 3rd-party audits are a ways from highest, however they provide a transparent trail for development.
The fourth theory is strong enforcement. Early ESG projects depended on reputational drive and stakeholder goodwill – issues that companies robotically not noted when income had been at stake. When alternate got here, it used to be as a result of laws established felony legal responsibility and monetary consequences.
Those penalties modified how companies assess possibility and proceed to form sustainability practices nowadays. Volkswagen’s 2015 ‘Dieselgate’ scandal, as an example, value the corporate over US$30 billion in fines, settlements and prison fees after U.S regulators discovered that the carmaker used to be dishonest emissions exams. BP confronted billions in consequences and liabilities for the 2010 Deepwater Horizon crisis, the most important oil spill within the historical past of marine oil drilling operations.
The present enforcement hole in AI creates a predictable dynamic. The anticipated price of AI washing – like doable funding good points, aggressive benefit, and marketplace valuation will increase – a ways exceeds the predicted value when it comes to consequences and possibility of detection. Till enforcement imposes penalties that exceed advantages, AI washing will persist as a rational industry technique somewhat than a possibility to a industry’s popularity.
Thankfully, buyers are starting to step up. The Federal Business Fee, as an example, introduced Operation AI Comply in 2024, focused on misleading AI claims, despite the fact that this effort has been in part scaled again by way of the present Trump management.
New requirements for a brand new technology
Till companies deal with those 4 ideas, AI washing will proceed. With out requirements and audits, even well-intentioned corporations can’t know if their paintings meets ok rigor. With out checks of subject matter affect, some teams of customers or shareholders shall be harm. And with out legal responsibility, even thorough auditors gained’t be capable to establish whether or not a industry’s claims about AI are fair.
Those ideas, carried out extensively, additionally assist give an explanation for why greenwashing persists. As an example, the loss of common reporting requirements continues to go away some gaps, with data-quality problems persisting at the same time as reporting frameworks emerged. Extra essentially, political buy-in for ESG has reduced considerably, in particular within the U.S., the place over 150 expenses had been offered throughout more than one states by way of 2023 to disincentivize companies from adopting ESG. Primary monetary establishments – together with JP Morgan, State Boulevard, BlackRock and PIMCO – have retreated from their previous local weather commitments amid political drive in addition to antitrust issues.
This pattern presentations that even properly designed duty measures require sturdy political reinforce to prevail. In the end, company sustainability took greater than 25 years to broaden from an preliminary framework to necessary requirements, and it nonetheless stays a piece in development. AI, in contrast, is advancing exponentially when it comes to its succeed in and societal affect. There might not be 25 years to catch up – however no less than there are courses from the new previous.
