Tech
Duolingo now has more than 100 million monthly active users
Sherwood News

Duolingo notifications: persistent, annoying, and oh-so-effective

The language-learning app now has more than 100 million monthly active users

Dua Lipa. Brazilian butt lifts. Clapping back in the comments.

Not a slam poem about the digital age, but rather some of the unconventional marketing strategies employed by the world’s leading mobile learning platform… by way of a green owl mascot with a penchant for intimidation.

Owl for one...

Duolingo — the app that encourages users to learn languages by completing short daily lessons — has exploded in popularity since its founding in 2011. In that time, it’s built a uniquely offbeat online presence, as well as a reputation for persistent notifications and overt gamification.

The company’s latest earnings, released Wednesday, reported its 5th consecutive profitable quarter, with total bookings of ~$190 million and net income of over $24 million, a more than 6x increase on the same period a year prior. It also saw Duolingo hit a significant milestone: 100 million monthly active users (MAUs).

Duolingo now has more than 100 million monthly active users
Sherwood News

Although the Q2 report sent Duolingo shares up more than 10% on the news, it’s been a bumpy year for the stock, which is still down 21% year-to-date. Analysts have pointed to a deceleration in growth for its vital daily active user count in the previous quarter as a reason why investors might be tentative to go all-in on lingo.

…and one for Owl

With users as its lifeblood, retaining aspiring polyglots, and eventually converting them to paying customers, is critical. That’s a tactic that Duolingo has down to a fine art, thanks to its 'make learning fun' approach to language tuition and heavily gamified user interface.

From the perspective of a Duolingo novice, the outline of how “studying” works on the app might seem more fitting to an arcade than a learning platform: users gain XP (experience points) as they learn, they can win gems, they keep hold of hearts, and, of course, they must return to the app day-after-day if they are to maintain their all-important streak.

It’s perhaps no surprise, then, that Duolingo is beloved by video game-raised Gen Z. For all its addictive hooks, though, the company’s quirky social media page is what’s really cemented the brand’s identity in the zeitgeist.

Duolingo’s TikTok follower count is soaring
Sherwood News

Sharp-tongued

Although it’s existed as a language-learning platform for over a decade, Duolingo’s foray into “unhinged” marketing began in 2021, when a 23-year-old graduate asked if they could make videos for the brand’s inactive TikTok account.

Since then, Duolingo has developed an online presence that at once straddles being timely, suggestive, and threatening — which younger audiences seemingly can’t help but respond to, with the company stating in a 2022 blog post that ~60% of all its US learners were under 30. As such, Duolingo’s TikTok followers continue to boom with its eccentric content, growing by 76% in the past year alone.

A masterclass in the marketing power of direct audience engagement and the cringe-inducing phenomenon of “trendjacking” (see: Duolingo chiming in on “brat summer” for some reason), Duolingo uses its image to push its primary objective, which ultimately pays its bills: getting people to use the app every single day.

Remember me?

This year’s Super Bowl wasn’t short of weird ads, but none were stranger than a five-second animation of a green owl producing its own face from its backside with the adage: “do your duolingo”.

If that wasn’t enough, the commercial was accompanied by a push notification for its US app users stating, “No buts, do a lesson now.” This kind of reminder was likely met with little surprise from Duolingo veterans, who are now used to being scolded by what Kelli María Korducki at Business Insider recently described as “the world’s meanest app”.

Indeed, people who download Duolingo for the sake of learning a language are often taken aback by the blackmail-adjacent tactics used by the app to keep them on-track. If you’re completing daily lessons, you will receive encouragement (“Your health is full again!”). But, if you’re inactive for too long, Duolingo’s messages start to sour: “You’re falling behind!”; “These reminders don’t seem to be working”; “It looks like you've learned how to say 'quitter' in Portuguese."

Still, whatever anyone might think of its “evil Duolingo owl” persona and broader marketing tactics, they seem to work.

Duolingo conversion rates are rising
Sherwood News

Pushed for cash

Turns out, Duolingo’s guilt-tripping is getting results. In the company’s most recent quarter, ~33% of MAUs were reported as using the app daily, up nearly a full 10 percentage points on the same quarter 5 years ago. The proportion of MAUs that pay for the service has more than doubled over the same period.

The successful conversion of ad-tiered to paying users makes Duolingo one of the most effective freemium models in the industry. And, beyond its ad-free, $12.99-monthly Super Duolingo tier is an even more advanced service: Duolingo Max, which incorporates AI-generated features for an extra $17 a month.

Language model

Indeed, Duolingo was one of the first major customers of OpenAI’s GPT-4, and since March 2023 it has rolled out “highly-personalized language lessons” using the technology, including chatbots and an “explain my answer” feature. In January, the company cut 10% of its contractor jobs, which was blamed at least in part on the use of AI.

It’s impossible to predict what language learning might look like in years to come, and perhaps highly gamified, AI-driven apps will be the (somewhat depressing) future. But, at least we don’t have to wonder what it would be like if the Duolingo bot becomes so advanced that it develops feelings — the owl already hates you for missing your lessons.

More Tech

See all Tech
tech

Report: Amazon’s AI bots have been behind multiple AWS outages

Amazon’s AI tool Kiro, which launched in July and can code autonomously, was behind a 13-hour interruption to Amazon Web Services in December, according to reporting by the Financial Times.

The FT reports that the company’s AI tools have caused AWS service disruptions at least twice in recent months.

In the December outage, which Amazon called an “extremely limited event” that did not have an impact on customer-facing service, engineers allowed Kiro to make changes and the tool opted to “delete and recreate the environment.”

Amazon has a closely tracked internal target that 80% of its developers use AI to code once a week, employees told the FT. The company says the December incident was a “user access control issue” and not an issue with Kiro’s permissions.

AWS accounted for 57% of Amazon’s operating profit in 2025. In December, following a larger outage months earlier, AWS and Google announced a partnership to attempt to prevent massive network outages.

Update, February 20, 5:50 p.m. ET: In a statement to Sherwood News, an AWS spokesperson disputed the report, writing:

“These brief events were the result of user error—specifically misconfigured access controls—not AI. The December service interruption was an extremely limited event when a single service (AWS Cost Explorer—which helps customers visualize, understand, and manage AWS costs and usage over time) in one of our two Regions in Mainland China was affected. This event didn't impact compute, storage, database, AI technologies, or any other of the hundreds of services that we run. We are not aware of any related customer inquiries resulting from this isolated interruption. Following these events, we implemented numerous additional safeguards, including mandatory peer review for production access, enhanced training on AI-assisted troubleshooting, and resource protection measures. Kiro puts developers in control—users need to configure which actions Kiro can take, and by default, Kiro requests authorization before taking any action.”

In the December outage, which Amazon called an “extremely limited event” that did not have an impact on customer-facing service, engineers allowed Kiro to make changes and the tool opted to “delete and recreate the environment.”

Amazon has a closely tracked internal target that 80% of its developers use AI to code once a week, employees told the FT. The company says the December incident was a “user access control issue” and not an issue with Kiro’s permissions.

AWS accounted for 57% of Amazon’s operating profit in 2025. In December, following a larger outage months earlier, AWS and Google announced a partnership to attempt to prevent massive network outages.

Update, February 20, 5:50 p.m. ET: In a statement to Sherwood News, an AWS spokesperson disputed the report, writing:

“These brief events were the result of user error—specifically misconfigured access controls—not AI. The December service interruption was an extremely limited event when a single service (AWS Cost Explorer—which helps customers visualize, understand, and manage AWS costs and usage over time) in one of our two Regions in Mainland China was affected. This event didn't impact compute, storage, database, AI technologies, or any other of the hundreds of services that we run. We are not aware of any related customer inquiries resulting from this isolated interruption. Following these events, we implemented numerous additional safeguards, including mandatory peer review for production access, enhanced training on AI-assisted troubleshooting, and resource protection measures. Kiro puts developers in control—users need to configure which actions Kiro can take, and by default, Kiro requests authorization before taking any action.”

$830B

OpenAI is finalizing commitments on a funding round that could climb beyond $100 billion at a valuation of $830 billion, according to a report from The Information.

Per The Information, SoftBank is expected to invest $30 billion into the ChatGPT maker, spread across the year in three installments of $10 billion. Up to $50 billion could come from Amazon and $30 billion from Nvidia (up from the $20 billion Bloomberg reported earlier this month). An additional investment in the low billions could come from Microsoft.

OpenAI was last valued at $500 billion following a fundraising round completed in October. Earlier this month, its rival Anthropic took in $30 billion from investors including Microsoft and Nvidia at a $380 billion valuation.

tech

Tesla’s 45 Austin Robotaxis now have 14 crashes on the books since launching in June

Since launching in June 2025, Tesla’s 45 Austin Robotaxis have been involved in 14 crashes, per Electrek reporting citing National Highway Traffic Safety Administration data.

Electrek analysis found that the vehicles have traveled roughly 800,000 paid miles in that time period, amounting to a crash every 57,000 miles. According to the NHTSA, US drivers crash once every 500,000 miles on average.

The article says Tesla submitted five new crash reports in January of this year that happened in December and January. Electrek wrote:

“The new crashes include a collision with a fixed object at 17 mph while the vehicle was driving straight, a crash with a bus while the Tesla was stationary, a collision with a heavy truck at 4 mph, and two separate incidents where the Tesla backed into objects, one into a pole or tree at 1 mph and another into a fixed object at 2 mph.”

Tesla updated a previously reported crash that was originally filed as only having damaged property to include a passenger’s hospitalization.

Last month, Tesla shares climbed after CEO Elon Musk said in a post on X that the company’s Austin Robotaxis had begun operating without a safety monitor.

The article says Tesla submitted five new crash reports in January of this year that happened in December and January. Electrek wrote:

“The new crashes include a collision with a fixed object at 17 mph while the vehicle was driving straight, a crash with a bus while the Tesla was stationary, a collision with a heavy truck at 4 mph, and two separate incidents where the Tesla backed into objects, one into a pole or tree at 1 mph and another into a fixed object at 2 mph.”

Tesla updated a previously reported crash that was originally filed as only having damaged property to include a passenger’s hospitalization.

Last month, Tesla shares climbed after CEO Elon Musk said in a post on X that the company’s Austin Robotaxis had begun operating without a safety monitor.

tech
Jon Keegan

Ahead of IPO, Anthropic adds veteran executive and former Trump administration official to board

Anthropic is moving to put the pieces in place for a successful IPO this year.

Today, the company announced that Chris Liddel would join its board of directors.

Liddel is an seasoned executive who previously served as CFO for Microsoft, GM, and International Paper.

Liddel also comes with experience in government, having served as the deputy White House chief of staff during the first Trump administration.

Ties to the Trump world could be helpful for Anthropic as it pushes to enter the public market. Its reportedly not on the greatest terms with the current administration, as the startup has pushed back on using its Claude AI for surveillance applications.

Liddel is an seasoned executive who previously served as CFO for Microsoft, GM, and International Paper.

Liddel also comes with experience in government, having served as the deputy White House chief of staff during the first Trump administration.

Ties to the Trump world could be helpful for Anthropic as it pushes to enter the public market. Its reportedly not on the greatest terms with the current administration, as the startup has pushed back on using its Claude AI for surveillance applications.

tech
Rani Molla

Meta is bringing back facial recognition for its smart glasses

Meta is reviving its highly controversial facial recognition efforts, with plans to incorporate the tech into its smart glasses as soon as this year, The New York Times reports.

In 2021, around the time Facebook rebranded as Meta, the company shut down the facial recognition software it had used to tag people in photos, saying it needed to “find the right balance.”

Now, according to an internal memo reviewed by the Times, Meta seems to feel that it’s at least found the right moment, noting that the fraught and crowded political climate could allow the feature to attract less scrutiny.

“We will launch during a dynamic political environment where many civil society groups that we would expect to attack us would have their resources focused on other concerns,” the document reads.

The tech, called “Name Tag” internally, would let smart glass wearers identify and surface information about people they see with the glasses by using Meta’s artificial intelligence assistant.

Now, according to an internal memo reviewed by the Times, Meta seems to feel that it’s at least found the right moment, noting that the fraught and crowded political climate could allow the feature to attract less scrutiny.

“We will launch during a dynamic political environment where many civil society groups that we would expect to attack us would have their resources focused on other concerns,” the document reads.

The tech, called “Name Tag” internally, would let smart glass wearers identify and surface information about people they see with the glasses by using Meta’s artificial intelligence assistant.

Latest Stories

Sherwood Media, LLC produces fresh and unique perspectives on topical financial news and is a fully owned subsidiary of Robinhood Markets, Inc., and any views expressed here do not necessarily reflect the views of any other Robinhood affiliate, including Robinhood Markets, Inc., Robinhood Financial LLC, Robinhood Securities, LLC, Robinhood Crypto, LLC, or Robinhood Money, LLC.