Unbabel has now distanced rival language industry startups such as Lilt, Smartcat, or, more recently, lengoo, by a considerable margin in terms of amount raised (although the exact amount raised by DeepL in a Benchmark Series A remains unknown).
With this latest funding round, Unbabel has arguably become the most prominent example in the language industry of what Redpoint Managing Director and SlatorCon San Francisco panelist Tom Tunguz has called the “AI Agency” — a new kind of company that is using the latest advances in machine learning “to disrupt a market dominated by agencies.”
Laser-Focused
Over the past two years, Unbabel has been laser-focused on building neural machine translation-based services and technology to enable multilingual customer service. This means Unbabel’s clients tend not to be the typical enterprise localization buyers.
In an email interview with Slator, Unbabel CEO Vasco Pedro said that as their “current focus is on multilingual customer service, we interface with the customer service heads, as they will directly benefit from a more efficient and scalable CS operation. We also see interest from executives in charge of digital transformation that are looking at how technology and new business models can positively impact their business. We do see more technology and forward-looking localization departments being involved and even championing our approach as this is the first time they feel they can bring value to use cases that need speed and scale in the translation process.”
While longer-term, there is a “much bigger vision,” according to Pedro, multilingual customer service will remain Unbabel’s focus for now. “Our technology is evolving every day and we’re already piloting with video and chatbots, although it’s too early to say they will be our next use case,” the CEO said.
Right Kind of Industry Dynamics
While at the time of SlatorCon San Francisco, Point72 Ventures’ investment in Unbabel was not yet public, Sri Chandrasekar, who co-leads the fund’s investments in AI and machine learning, pointed to customer service as an attractive market for language services.
At the time, Chandrasekar explained why the language industry appeared on his firm’s radar in the first place: “I think it’s a combination of the right kind of industry dynamics. It is a big industry, there is a lot of labor, and a lot of amazing big customers who have a high willingness to pay. There’s also a technology chain that should enable new kinds of business models and an expansion of the types of opportunities that players can go after.”
At the heart of Unbabel lies the thesis that combining state-of-the-art neural machine translation with a human-in-the-loop production model will result in a hyper-competitive offering for professional-quality translation at a much lower price point than is possible today.
Unbabel’s translator pool consists of both professional and non-professional linguists. “They don’t have to be qualified translators,” CEO Pedro said. “Leveraging our AI technology, we’re able to have non-professional translators to produce professional-grade translations for the use cases we currently support.”
Slator Machine Translation Expert-in-the-Loop Report
60-page report on the interaction between human experts and AI in translation production, including AI-enabled workflows, adoption rates, postediting, pricing models.
Hiring Challenge
A significant portion of the funds raised will be invested in building out Unbabel’s AI team. In April 2019, Unbabel hired Amazon’s former Machine Translation Group Manager Alon Lavie, who is now building the startup’s AI research lab in Pittsburgh. The city in Pennsylvania is, of course, home to Carnegie Mellon University, which runs one of the world’s leading graduate programs in artificial intelligence.
Being backed by leading VCs certainly helps attract top talent in the highly competitive market for AI researchers and engineers. Pedro said Unbabel has a “good combination of still being small (compared to big-tech) but showing consistent growth — and that creates a great opportunity for great talent to have a relevant impact when they join the team.” According to Pedro, Unbabel current employs over 230 full-time staff.
Bullish Outlook
VC investor Chandrasekar is bullish on the language industry’s future. Asked by a SlatorCon participant about the impact of unit prices dropping as automation advances, Chandrasekar responded: “I absolutely think that over the course of the next five years, there will be a tremendous amount of pricing pressure in the language translation market [in areas such as] high-quality translations for healthcare, financial services, and legal, etc. But I think it is going to be more than made up for by the fact that technology is going to make possible a whole lot of new cases. As a result, the best companies are not going to shrink, they are actually going to grow, because they suddenly are able to deliver far more to an enterprise customer than what they were delivering before.”
Photo: Unbabel CEO Vasco Pedro
