Deceptive AI Marketing: The new falsehood from Silicon Valley
“Could you please review all the historical pitch presentations and replace ‘cryptocurrency’ with ‘Artificial Intelligence’?
“Could you please review all the historical pitch presentations and replace ‘cryptocurrency’ with ‘Artificial Intelligence’?” This heading from a cartoon in The New Yorker by Benjamin Schwartz perfectly encapsulates the latest trend of deceptive AI marketing in Silicon Valley.
Perception of AI marketing may seem like just another spin tactic, but it’s actually a intricate and multi-faceted occurrence. It is crucial for all readers of this piece — including tech innovators, advertisers, product developers, users, and IT specialists of all types — to comprehend the exaggeration, distorted focuses, and blatant dishonesty that we come across not only in promotional and sales strategies but also in the narratives we consume based on industry assertions.
Understanding the Deception in AI Marketing
AI washing is a misleading advertising technique that exaggerates the significance of artificial intelligence in the product or service being endorsed. The concept is inspired by “greenwashing,” introduced by environmental advocate Jay Westerveld in 1986, where consumer goods are promoted as eco-friendly irrespective of their environmental consequences.
Products employing traditional algorithms are branded as “AI-driven,” exploiting the lack of a universally accepted delineation for what constitutes AI and what doesn’t. Startups create applications that connect to a publicly available generative AI API and market them as AI apps. Large-scale AI projects meant to showcase technology often rely on human intervention behind the scenes, as humans are crucial to making the intricate AI solution operational.
Now, focusing on the final point…
AI: The Human Element
E-commerce behemoth Amazon unveiled 44 sophisticated outlets under the Amazon Go and Amazon Fresh brands, utilizing the company’s “Just Walk Out” suite of technologies starting in 2016. Amazon’s vision was to offer stores where customers could enter, select items from shelves, and exit without encountering a cashier. Sensors, such as cameras, fed data to AI systems, which could identify purchasers and bill them accordingly — all without the need for a traditional checkout process. It gave the sensation of shoplifting, within legal bounds.
The system operated with cutting-edge computer vision technology that monitored customers and their chosen products. Sensors embedded in shelves gauged the weight of lifted items, validating the type and quantity of products recognized by the cameras. RFID-tagged items contributed additional information to the mix. Sophisticated machine learning algorithms processed data from cameras and sensors to recognize products and connect them to specific shoppers. Automated gates at entry and exit points identified individuals coming and going.
The algorithms underwent training on millions of AI-generated images and videos to recognize products, human actions, and behaviors.
For over seven years, Amazon was eager to expound on the components of its Just Walk Out technologies but remained reticent about the approximately 1,000 employees hired to facilitate its operational efficacy — only acknowledging their presence after media exposure. Even then, Amazon obscured the specific functions of these employees, stating only that they were not tasked with video review.
Despite having 1,000 employees overseeing and supporting 44 stores (monitoring three-quarters of transactions on average, according to reports), the technology has encountered challenges, such as delayed billing, mishandled orders, and elevated operational costs.
In the present year, Amazon has been phasing out the utilization of Just Walk Out technology in its primary establishments but continues to offer it as a service to external entities.
Another notable instance of human involvement behind the AI façade is the realm of autonomous vehicles.
Google’s Waymo division (previously Google’s self-driving car initiative) maintains a control center reminiscent of NASA where personnel monitor vehicles through cameras and intervene remotely in case of issues. General Motors’ Cruise subsidiary acknowledged that their self-driving taxis require human intervention every 4 to 5 miles on average, with each remote handling session lasting around 3 seconds.
Many other self-driving companies lean even more heavily on remote human operators. A German firm named Vay straightforwardly utilizes remote human operators to maneuver their vehicles. The company recently introduced a valet parking service in Las Vegas, where a car is remotely driven to the user, who then takes control until reaching their intended destination, after which a remote operator parks the vehicle.
Amazon’s stores and self-driving cars are just glimpses of a prevalent trend.
Reasons for the Prevalence of AI Marketing Deception
The high-ranking, well-compensated technologists fabricating AI systems firmly believe in the potential of AI to resolve exceedingly intricate problems — which in theory, it can. They assert to superiors that it’s achievable. Those leaders relay this to the board. C-suite executives assure investors of the feasibility. Consequently, as a corporate body, they assure the masses that it’s a reality.
But there lies a significant hitch: it’s not feasible.
Most enterprises feel compelled to justify lofty assertions and thus conceal the extent to which their product or service relies on human intervention behind the scenes to tackle issues, negotiate through challenges, and facilitate the semblance of technological prowess.
Unabashed companies remain unfazed by concrete evidence that their AI capabilities fall short of their claims, perpetually renewing their assertions. Tesla’s CEO Elon Musk is a prominent example of this.
Back in October 2016, Musk pledged that Tesla would display a fully autonomous journey from Los Angeles to New York by the close of 2017.
By April 2017, he speculated that within about two years, passengers would be able to nap in their vehicles while it autonomously navigated.
In 2018, Musk postponed the fully self-driving Tesla target to the end of 2019.
Come February 2019, he guaranteed full self-driving capabilities “this year.”
In 2020, Musk declared the deployment of over 1 million self-driving robotaxis by year-end.
Even in the current year, Musk insinuated that fully self-driving Teslas might be feasible “later in the year.”
Reality check: It’s unlikely to materialize. Musk is misleading both himself and his patrons. Musk is the quintessential face of AI deceit.
The Detrimental Implications of AI Marketing Deception
The collective effect of AI deception is that it leads the public and the tech sector astray, propelling the illusion that AI can achieve feats it’s incapable of. It fosters the misconception that AI serves as a universal remedy for all challenges or a pathway to dystopia, contingent on one’s perspective.
AI deception incentivizes subpar solutions, fixating on perceived “wizardry” rather than merit. The assertion that your pet grooming apparatus is “AI-powered” doesn’t translate to a cleaner pet; it merely signifies an overpriced grooming tool.
AI deception distorts funding strategies. Present-day Silicon Valley investments are engulfed by genuine AI and AI-deception projects. Even astute investors may overlook AI-deception falsifications and exaggerations, recognizing that the AI narrative will ensnare buyers owing to their gullibility.
Nonetheless, the most pressing issue is not the deceptive practices within the industry but self-deception. Purveyors of AI solutions view human involvement as a blemish when in reality, I believe that human participation should be viewed as a relief.
In truth, consumers desire human interaction in their shopping and commuting experiences.
What’s truly required is more human touch and less reliance on automation. Presently, the digital sphere is inundated with poor quality content generated by AI, alongside bizarre, sometimes unsettling visuals. Google’s zealous drive to supplant its search engine with an answer engine results in absurd outcomes, like adhesive on pizza.
What the public genuinely craves is a search mechanism directing us toward human-crafted content or, at the very least, a ranking system that prioritizes human-created content over AI-produced material.
AI-deception thrives on falsehoods. It’s predicated on the notion that individuals prefer machinery to control and orchestrate every facet of their lives, which isn’t accurate. It’s built on the misconception that integrating AI inherently augments functionality, which isn’t the case. And it’s built on the fallacy that engaging human resources signifies a deficiency in technology, which is far from the truth.
Enough with the delusion of AI deception! Sellers must be forthright about the role of AI. Buyers must demand evidence that any AI featured in the products or services they invest in delivers tangible value.
On behalf of the technology sector, the tech consumer community, and the tech journalism realm, I implore Silicon Valley: Cease the charade of deceptive AI marketing.
About Author
