When Duolingo CEO Luis von Ahn in April announced a dramatic pivot of his popular language learning app to being entirely “AI-first,” it landed the small company in the headlines. Von Ahn’s letter to his staff explained he was conscious that “AI is already changing how work gets done. It’s not a question of if or when. It’s happening now.” He was essentially hoping to get ahead of the trend, and as well as using AI to generate content for the app’s language-learning users, the shift would also mandate the company’s workers “rethink much of how we work” including gradually shifting away from using contractors to do typical duties that AI can now handle. 

Von Ahn’s big AI push seemed like a sensible, pragmatic move, at least on paper. But it generated extraordinary pushback from customers. Newspaper the Financial Times gathered together some comments: “Unbelievable,” said one user on LinkedIn, “Cancelling my account NOW,” “Well, there goes my 1,098 day streak,” posted another. But it was more than this. The CEO’s announcement triggered tough questions about the future of work, education, and the direction of the company brand. It also created a self-inflicted public relations crisis for the company, which successfully gamified language learning and developed a loyal customer base. Customers reported cancelling subscriptions in droves.

The Times spoke to von Ahn, and he explained he feels many social media commentators actually misunderstood the “all-in” AI push to mean he was firing all his staff, and pivoting to a model where “everything is being controlled by a massive AI.” This tapped in to public “anxiety” about the impact of AI on the workplace, potentially displacing real humans from their jobs—a notion that’s resulted in many worrisome headlines over the last year. Von Ahn said he really should’ve “been more clear to the external world” when explaining his AI push. He felt that one of his big slip-ups was being “open about it,” even though “every tech company is doing similar things.” 

Despite the negative public reaction, von Ahn says his own staff got what the plan was and “nobody misunderstood,” because everyone knew Duolingo was “going to be a technology-forward company.” The Times said the most questions staff had about the changes was about the idea their performance reviews were going to include checking on their AI use. Here Voh Ahn’s intentions echo other pro-AI leaders, including drug maker Moderna’s CEO Stephane Bancel said last year he expected his staff to be consulting ChatGPT at least 20 times a day. 

The backlash has been so ferocious that von Ahn has had to make public explanations in an effort to clarify Duolingo’s path forward. He’s made a point of saying only a “very small number of hourly contractors who are doing repetitive tasks that we no longer need” will be impacted, and that “many of these people” will be offered contract work doing “other stuff.” The company is also still recruiting real people, if the role they’ve being considered for can’t be automated.

He also said that younger staff the company’s bringing on board are already exhibiting a “different mindset” because they’ve extensively used AI already – including at university, which echoes recent reports that basically every college student is busy using AI to “cheat” their way through their studies, which has triggered worry about how this will impact the world of work. 

Von Ahn used many similar arguments about AI as other proponents of the increasingly disruptive technology. The shift to turning some functions over to AI is really all about speed and efficiency, and von Ahn said Duolingo is looking at using it most in areas like writing code or generating the “very specific style” of images the company uses in its products. Artists won’t be scrambling over details, and instead will act like “more of a creative director” as they supervise AIs through the tasks, he said.

Von Ahn’s walking back of his AI pivot has driven yet more criticism online, since it smacks of a CEO in panic. In a discussion on Reddit, one user nailed it: “They’ve been getting absolutely cooked on every social media platform I’ve seen them on. Not sure how that’s impacted users, but it has absolutely caused them trouble with marketing. I’m not surprised to see him walk back the hardline stance, though I’m curious to see if it’s just lip service or an actual policy change.” Another commenter described the brand damage neatly: “They were one of the most well liked companies online ever. They captured the genz audience perfectly on TikTok and had several viral videos every week and presence on the comment section comparable to the most popular influencers. And it’s all gone.”

Why should you care about this? It’s just a very public struggle by one small app-making company, after all. 

You should care because it’s a cautionary tale about AI adoption, with several lessons for your company to consider. When embracing AI tools, you should be clear to customers and staff about the impact it’s going to have on the workplace. And you should be conscious that many members of the public remain very suspicious of AI, and if you brand your products as being made, even in part, by AI, you may expect some pushback. – Inc./Tribune News Service