Duolingo, Other Firms & Users Double Down On AI Tech Despite Diverse Warnings

As the capabilities, applicability, and use cases of artificial intelligence (AI) and generative artificial intelligence (GenAI) keep improving, the adoption by companies, consumers, and users will only keep increasing.

Considering how AI carries characteristics such as boosting productivity, reducing costs, improving turnaround time, streamlining processes, boosting operational efficiency, and more, it’s no surprise that businesses are opting to integrate the technology throughout their workflows, workforce, and working methods.

It doesn’t matter which industry a company is engaged in; they’re all looking to AI.

This includes Duolingo, an educational technology company that offers over 43 certificate language courses, spanning English, Spanish, French, and more, including less commonly studied languages (Navajo) and constructed languages (Klingon).

Through its LinkedIn handle, Duolingo published an all-hands email from its CEO and Co-founder, Luis von Ahn, which outlined the company’s AI strategy going forward—the gist of which was that Duolingo is going to be AI-first.

“AI is already changing how work gets done. It’s not a question of if or when. It’s happening now. When there’s a shift this big, the worst thing you can do is wait,” reads an excerpt from the email.


The email also mentions that Duolingo was one of the first companies to build a mobile-first application, which helped it win the 2013 iPhone App of the Year.

“Betting on mobile made all the difference. We’re making a similar call now, and this time the platform shift is AI.”

The move comes as Duolingo believes AI isn’t just a productivity boost but instead will help it get closer to its mission. Teaching requires a massive amount of content, which poses challenges when manually scaled. As such, the company turned to AI to replace “a slow, manual content creation process” with AI—one of its “best decisions.”

AI also helped the company build its Video Call feature, which was “impossible to build before.”

Despite Duolingo having to rethink much of how it works, the move is set to go ahead and will see the company roll out “a few constructive constraints.” This includes gradually switching contractors for AI, AI playing a key role in job profiles and performance review evaluations, headcounts coming in only if teams can’t automate more work, and functions possessing specific initiatives to align towards AI.

At the same time, the company promises to “remain a company that cares deeply about its employees” and says the move isn’t about replacing employees with AI but is about removing bottlenecks.

“Being AI-first means we will need to rethink much of how we work,” reads the email. “In many cases, we’ll need to start from scratch. We’re not going to rebuild everything overnight, and some things-like getting AI to understand our codebase-will take time. However, we can’t wait until the technology is 100% perfect.”


It’s not just businesses leveraging AI tech that are getting ahead. Even the makers of AI are succeeding.

Google’s advertising revenue was 8.5% more than what it expected it to be in Q1-2025, leading to its parent company, Alphabet, witnessing a 3% rise in its share value, all thanks to its bold bets on AI.

Ahead of this, even individual AI users are making big bucks.

Gen Z investors—including teenagers—are using GenAI chatbots such as ChatGPT to learn about the stock market and invest in it. Moreover, they’re making a good amount of money through this tactic.

According to the 2024 Charles Schwab Modern Wealth Survey, Gen Zs are also investing in stocks a lot sooner than previous generations did—on average, Gen Z is getting in by the age of 19, in comparison to Millennials at 25 and Gen X at 32.

As most in the world are flourishing thanks to AI, experts are still sounding sirens about its downsides and potential dire consequences.

Geoffrey Hinton, who’s known as one of the “Godfathers of AI,” has reiterated his stance that the rapid and unprecedented growth of AI, coupled with its lightly regulated nature, poses a threat to humanity. Hinton believes that people don’t understand what’s coming, and that there’s a 10 to 20 percent chance of AI eventually taking control away from humans.

Concerns related to AI tech have been echoed by Apollo Research, too.

In their “AI Behind Closed Doors: a Primer on The Governance of Internal Deployment” report, the AI evaluations startup addresses the absence of governance related to the internal deployment of highly advanced frontier AI systems.

Through the report, the firm recommends the implementation of overarching controls, including scheming detection and control, internal usage policies, oversight frameworks, targeted transparency, and disaster resilience plans.

It also covers topics such as the scope and urgency of Internal deployment, its characteristics and scenarios of concern, existing and proposed frameworks in internal deployment and across other safety-critical industries, and recommendations for internal deployment governance.

The startup finds that automated AI-powered research and development (R&D) is a core risk factor of misalignment. Here, AI R&D refers to the use of AI systems to design, train, or improve other, in most cases, more novel AI systems.

AI companies that develop frontier AI models are likely to deploy these tools across their organization. This ecosystem is currently ungoverned, since it’s internal. Furthermore, the models could also be used to automate a variety of tasks, including extended research and development (R&D). This could unintentionally lead to the models circumventing internal guardrails, too, allowing them to carry out a variety of destructive actions.

Moreover, as they become more powerful, they could be used to train and better themselves, eventually running in a self-reinforcing loop that sees the AI keep building better versions of itself, resulting in systems that are beyond oversight, uncontrollable, or result in undetected power accumulation.


Do you think the internal deployment of AI tools needs to be regulated just as much as public offerings?

Let us know in the comments below!