Arne Schepker, who has served as the chief executive of the Berlin-based language learning company Babbel, will step down and Markus Witte, co-founder and former chief executive, will take the reins to guide the company into a "new phase while they are searching for Arne's successor with patience," the company said. No surprise: this new phase will include AI.

Instead, Witte will not take the post of CEO now but will be the executive chairman and managing director.

Nearly exactly five years ago, Schepker became the only CEO; and a few months earlier co-CEO alongside Witte, who had been the CEO of the previous years. He told me he declined to have his contract renewed. Witte stays in the position of Executive Chairman and additionally becomes Managing Director.

I just couldn't get to a strong enough 'yes' and as a CEO, I don't feel you can do the job with just 100 percent commitment. It needs to be 180 percent, no matter what. And I couldn't get there, and I didn't feel that was enough and right and sufficient for the team, for the company, for our shareholders, so I decided not to extend my contract," he said.

In 2015, he joined Babbel as its first CMO. By now, he says, he's starting to see repetition. He just couldn't work up the enthusiasm for one more iteration of building an annual budget and establishing OKRs for the organization.

"That's reason number one. Reason number two is that the timing actually is quite good, because we are, anyways, moving into a new phase as a company," he said. In addition, he also wants to take more time to travel with his family for at least the next year. "Don't expect anything on my LinkedIn feed until next winter."

He also added that he was glad to do our interview with Witte as well together. "I believe that a founder who himself built the company, built our first products, built our culture, built all the foundations that I was able to work off of, and someone who I deeply trust and have a strong alignment with is an amazing transition," he said.

Babbel's revenue grew sixfold to about $300 million with nearly a 1,000-member team.
"We have done what we wanted to do." But he is perhaps most proud that the company was able to service those students' needs free-of-charge during the pandemic and now is doing for Ukrainian refugees.

"There's no monetary value to that. I can't even prove you an ROI on it. Still to this day, I cannot, but not a conversation goes by where that's not brought up, whether that's a press interview, a candidate interview, or just a dinner with friends."

So, what's the next phase of Babbel? According to Witte, "I believe that as the technology is changing, now AI can more directly play a role in helping people learn a new language.". Babbel always used machine learning under the hood, but it never billed itself as an "AI company." Rather, it always emphasized the teachers and experts it worked with to create its courses (in no small part to differentiate from competitors like Duolingo).

With the technology moving this fast, though, Witte also acknowledged that it's hard even to think about strategy more than half a year ahead.

"We're in a phase where even people building large language models don't know what the next generation will be able to do," he said. "And so I think even companies of our size, so not very early-stage startups, need to be more nimble than they ever have been."

And at this stage, he believes that having the founder return to the corner office is even to his advantage, as it is less difficult for him, being a founder-and the largest shareholder in the company-to make the high-risk overhauls to the company's strategy.

According to Witte, we are now there-the confluence of large language models, which are especially good at language-related tasks, and Babbel's deep expertise in language learning-can change the way the company goes about teaching its customers. Before, the technology wasn't there. "We have come to the point where what we said before is not true anymore," he said.

Schepker also noted that at its very core, Babbel's mission and the problem it aims to solve is a human one.

"It's still solving the problem of human language learning," he said. "You still want to speak to someone in another language. You want to have a conversation with a loved one, a family member, whatever it is.". "There's a real opportunity here for Babbel to use all the didactic knowledge that we have, to use all the data that we have, to use this new technology and put that together and create a real, personalized and powerful language learning journey that finally gets us to crack the problem for real. Because we've made language learning easier, but it's still work."

But aside from that, navigating change as AI already starts introducing, Witte says he is also keen on having more "moments of delight" on behalf of employees and users. "These things that bring a smile to your face on all different levels-that's what I'm driving towards," he said in our conversation. "That's kind of my mental model in the moment.". I don't think we must be great at everything. I don't think we need to perfect everything, but I do want these moments of joy on every dimension, on every level.