日本語 
Dieter, can you tell us about your professional background and your current role at Boostlingo?
My path feels like it has been anything but linear but then in some ways upon closer reflection it has become clearer to me that many of my career path choices make a lot of sense when put into the context of what has piqued my interest and excited me over the years. I’ve always been keenly interested in innovation and business that exists at the intersection of technology and communications, and this has largely guided my trajectory. I began in the film and media production industry in the San Francisco/Bay Area eventually becoming deeply involved in the burgeoning interactive CD-ROM and later Web media production space. I was fortunate to be there for several years during the peak of the dot com boom era. From there I moved to the east coast and completed a master’s degree in the Educational Psychology program at McGill University in Montreal. After graduation there I spent several years running global digital e-learning programs which included localization responsibilities for the simultaneous shipment of multilingual global education programs for several big-name tech companies. These roles and opportunities were a result of my good fortune to join an e-learning start-up in 1998 something which got me hooked on/addicted to the startup tech company vibe. We were eventually acquired by a larger e-learning company and that lead me to a variety of exciting roles in the language services and language technology sector. I spent many years in sales and marketing and international program management for enterprise language services. My time in this space was the most formative in developing a passion for real-time multilingual communications. Today I’m one of the co-founders of Boostlingo, where I am a brand evangelist and oversee our Global Strategy and Business Development activities including partnerships and strategic alliances as well as international growth strategy and delivery. Beyond this I try to contribute regularly to Boostlingo thought leadership and content development and delivery, including working on some of our research reports and via participation in regular webinars and conferences. I also host the Boostlingo podcast BoostCast we are in the second year of this. We started Boostlingo in 2016 with the goal to innovate, improve and redefine remote interpreting delivery, which had always been a challenge for me while working at previous language technology companies before Boostlingo. In my mind there was a better way to do it and an urgent need to democratize the technologies and tools to expand language access globally. Ten years on, we’re 3,500+ customers deep and we’ve evolved into an AI-forward language intelligence platform which includes human interpreting, AI interpreting, translation, scheduling, the works. Since 2022 Boostlingo has been a 5-time INC 5000 company and a 4-time INC. Best Places to Work USA company, which is a special distinction that fills my heart the most. One number I care about deeply each year is when we discover our eNPS score! I tend to gloat here because we consistently rank in the very highest percentiles and our employee retention rate is and has been for years been in the mid 90s. I do have a lot of long-haul travel each year as I split my time traveling through at least 5 continents each year. I spend a fair bit of time in Austin where we are headquartered and the west coast of the US where many of our technology and product folks are located. I also visit our development and sales teams and our strategic partners across Europe, APAC, and the MENA region. That said, I’m always happy to return to my adopted home by the beach near Sydney in Australia. All the globetrotting has kept me very honest about how differently this industry works around the world.
Which experience from those earlier chapters still shapes the way you think about technology and human communication today?
I would say my formative media production years, hands down. When you create film and video content, you learn very fast that the gear (technology) is never the main story, it is the humans on either side of the lens who bring the real magic. Something that I most certainly cling to as the evolution of AI tools and technologies evolve. My time building digital learning products also reinforces this for me. Content and what is being communicated only works when it lands in the learner’s actual context. I have connected these dots very purposefully throughout my career and it was especially component to the Boostlingo story. Effective interpreting involves the same contextual care. It is never simply just a language-transfer problem you need to solve. It’s a communication problem that includes context, trust, and timing issues. That’s part of the reason why I’ve never supported the “AI replaces interpreters” framing. Humans a very much in the loop and critical for effective multilingual communications. Every new technology will bring with it its own mix of detractors and occasional overexuberant pundits. The truth always exists somewhere in the middle. I’ve watched tape become digital become streaming become AI. The delivery layers and mechanisms change over time, but the human need underneath has always remained.
What is one belief you held when co-authoring “The Remote Interpreter” that experience has since made you rethink?
The first thing that comes to mind is probably my naivety around the whole process involving book writing and publishing. Nothing could have ever prepared me for the actual level of effort required to get a book out the door! The question is great because I have reflected a lot on this recently as it’s been more than 5 years since we began to write that book which released in the summer of 2023. When we wrote The Remote Interpreter, the fight was still around legitimacy proving remote interpreting could stand shoulder to shoulder with onsite. Back then I believed modality was the really the big question and issue. The thinking was if you nail the audio, the video, the protocols and the training then the rest will sort itself out, essentially prove itself. However, since then the very recent boom and proliferation of AI tools and technologies around the interpreting space I’ve readjusted my position. Our State of Interpreting Technology 2026 research showed that the market has quickly become multi-modal with hybrid phone, video, onsite, bilingual staff, and now AI augmented machine interpreting to boot. The modality is no longer the question all options are on the table now and accepted. The circumstances of each interpreting encounter dictate the modality. The main problem that we are still seeing is that half of our survey respondents still reported a case where an LEP individual needed an interpreter and didn’t get one. So, modality was never really the endgame for us it is still availability and fulfilment. The real frontier ahead of us is language orchestration and language ubiquity. This involves technology that can streamline the coordination of the right resource, right moment, right context. I guess we were somewhat prescient calling the Remote Interpreter book “Volume One”. I think the content still very much holds up well today however a “Volume Two” will likely become necessary soon and most certainly will need to lean hard into AI augmentation and the transformation of the interpreting space into an integral part of the concept of Language Operations (LangOps) and as part of a ubiquitous and automated orchestration layer for multilingual support.
What do outsiders most often misunderstand about what makes a language technology business succeed beyond a strong product?
There is a misconception that the product wins the deal. It doesn’t, it just gets you the meeting. The experience and the outcomes are what really matter. The language access and support industry runs on high stakes, regulated, deeply human moments. It happens in hospitals, courtrooms, and classrooms. Nobody buys interpreting software being focused just on feature sets and technical prowess. They are buying reliability, adherence to governance and regulatory compliance, and the trust and confidence that there’s someone qualified who will pick up the phone when it matters. The other thing outsiders often miss or don’t immediately understand is that this is very much an ecosystem business. There are Interpreters, LSP partners, technology providers, associations, regulators and buyers and they all talk to each other all the time. It’s a relatively small industry, although revenue wise it is much larger than for example the global film and music industries combined. It is also an industry with a with a very long memory. How you treat your interpreters shows up in your enterprise pipeline eventually. We’ve spent ten years focused on building and maintaining confidence and earning trust across the entire ecosystem, and I’d argue that that’s been worth more than any single feature that we’ve ever shipped.
What separates an organization that has accumulated interpreting options from one that has built a mature language access operation?
Options around your interpretation solution is a procurement conversation, and very check-box utilitarian. Maturity is when an organization is having conversations around operations and solution orchestration and reliable outcomes. Our 2026 research, which included 370-plus stakeholders across the ecosystem, made this point very plain. Nearly everyone selling interpreting services successfully is multi-modal now. Their access has widened and their delivery modalities have expanded, but as I mentioned before, half of the organizations that we spoke with are still reporting unmet interpreter needs. Yes, cost is still the most-cited pain, but the real measure of value begins when the conversation turns to interpreting coordination. The customers are most concerned about the automation of their scheduling process, around the consistency and quality of the interpretation, and about the efficacy around the admin holding the whole thing together. The accumulators just bought channels and stapled them together. The mature operations now deploy systems with clear rules for which modality fits which encounter, define the criteria for when AI is appropriate, have a clear escalation path to qualified professional humans in the mix, and can quickly access the data feeding back into those decisions. You spot the level of maturity in the performance numbers they can cite. How fast someone reaches an interpreter using their system, how consistently the quality holds, and how much staff effort it takes to keep it all running. It is not in the number of whizz bang features you bring to the table nor the length and names of your vendor list to achieve any of this. No one cares. Language is everywhere. It is component to the end user experience. It is an operational layer. It is a utility. You need to speak real data and outcomes.
What part of interpreting is the industry underestimating precisely because it’s difficult to automate?
I would say the perception, expectations and/or judgment in terms of how things are happening or should be happening around the process of interpreting. AI is getting genuinely good at converting language we can’t pretend otherwise. But skilled interpreters are doing something else in parallel that people often fail to take into count and it’s not an insignificant thing right? Human interpreters are reading that the patient nodded but didn’t understand, they are managing conversation turn-taking, they are flagging a cultural landmine before it goes off, they are knowing when to step outside the script because something in the room feels “wrong”. Interpreters are becoming the critical communications risk managers. That work is hard to automate because it isn’t in the transcript, and the industry underestimates it because it never shows up on an invoice, until of course it does, and that’s in the moment and on the day it’s missing. Our 2026 data reflected clearly where people land on this. Even among non-users of AI, about 42% said they figured in low-risk activities AI use is acceptable. However, that’s an operative extremely important concept to take seriously! Low risk! Low risk means that everyone intuitively knows why and where the professional interpreter humans belong or at least they should. It becomes very apparent quickly when things come off the rails.
Where does the real intelligence of the system need to sit — the model, the rules, or the human escalation layer?
The models will keep getting better and, frankly, increasingly interchangeable. The durable intelligence sits in the layer that decides the orchestration and governance around the model(s). The intelligence asks and solves for what’s this encounter? What’s the risk level? What’s the language pair? Is AI appropriate here and if the situation shifts, how fast can a qualified human interpreter come into play? That’s where trust gets built and lives. Only 16.8% of respondents in our 2026 report currently use AI interpreting, with another 15.9% still evaluating it. So, the appetite is real and growing but it remains conditional. People don’t trust a system because the model sounds fluent. They trust it if it can demonstrate that it knows its limits. Human escalation isn’t a fallback bolted on simply for the optics and peace of mind it’s a deliberate design feature. Arguably it is the design feature.
In international growth, what tends to matter more than companies initially expect?
Again, I think trust here is the strongest differentiator these days and in the age of AI it is increasing all the time as the single most important decision factor for end users. Buyers are paying close attention to how differently one system is built and how it operates from one market to the next. Compliance and data sovereignty are table stakes just as much if not more as the key requisite platform technologies are. Systems must show real governance and regulatory compliance you can map directly to the locale. At Boostlingo we’ve literally built regional regulatory field guides for exactly that purpose.
What also catches companies off guard is the relationship physics. In LATAM countries the relationship precedes the transaction. You show up in person, in the language, more than once, before anyone talks contracts. In most parts of APAC and in the MENA/Gulf region, credibility routes through partners, associations and government relationships and is far less reliant on sales and marketing approach. It can be very nuanced and specific from region to region, and you would be crazy not to include local partners in your go-to-market strategy. NAATI may mean nothing to an interpreting company in the US or Japan, but in Australia and New Zealand, NAATI credentialing informs and shapes your buyer expectations from day one. The classic mistake is (and has always been) exporting your own geo-centric go-to-market playbook and expecting it to translate. It doesn’t. It blows my mind how many organizations operating globally still don’t get this. You won’t scale trust with just your landing page. International business is by its very nature a highly localized activity.
What do you increasingly hear from practitioners, LSPs or customers that the broader technology industry still hasn’t understood?
From practitioners we frequently hear “We’re not afraid of AI we’re afraid of having no say in when and how it’s being used.” That comes up constantly. Interpreters aren’t debating whether the technology is impressive. They’re asking who decides appropriateness, what the escalation path is, and whether anyone building these tools has sat through a real medical encounter. From buyers, it’s a different version of the same message. They say, “hey, stop selling me yet another feature or function or channel and make what I already have work more reliably for me (and cheaper)”. The tech world does itself no favour by talking up replacement economics, especially when the people doing the work rightfully talk about risk, context and workflow. I am hopeful we will be able to close that gap and everything else gets a little easier. Right now, there is palpable level of hostility and friction but there is also boundless opportunity and a real roadmap towards true universal and ubiquitous multilingualism. Which is honestly a big part of the reason and motivation in why we publish our blogs and our research with regular cadence and run internally and externally produced webinars and podcasts every month. The requirement to advocate and educate around language access and language technology is perpetual.
Also Read: Interview with Hiro Yokoki, Chief Visionary Officer @ Amulet plus .G.K
Over the next five years, what change will matter most to the industry, and what role do you want to play in shaping it?
I see in the immediate future a transformational moment where language stops being a bolt-on service and becomes embedded infrastructure, a native layer inside the platforms where people already communicate. This includes health systems, contact centres, classrooms, events etc. Hybrid AI-human interpreting delivery will become the operating standard rather than a transition state. Orchestration and trust will become the prime differentiators, and unmet demand will ultimately start shrinking because the cost of low-risk access will drop toward near zero. This will free skilled professional human interpreters for the complex, context reliant and high-stakes work that only they can do. Faster, fairer, and more human, not less.
My role? Boostlingo’s role? Same as it’s always been. We will continue to expand and democratize language access at scale and develop systems that help break down communication barriers everywhere. We will build the partnerships, develop the integrations, strengthen the critical guardrails around safety and privacy, publish the data and innovate and iterate based on our findings and via our conversations and interactions with customers and end users. We’ll host, moderate and join the important conversations and causes, and we’ll keep pushing on what multilingual communication itself can mean. I am personally excited about new cutting edge delivery modalities like holographic presence interpreting, which we’ve already been exploring. I personally embrace the concept of “language is everywhere” aka language ubiquity will eventually mean language access in the internet of things. I want to be able to communicate effortlessly with my refrigerator or rental car in Barcelona or Osaka. I want to empower and interact with humanoid robots that can speak up to 7,000 languages Ten years in, I still think we’re so very very VERY early. Which is exciting. We’re really just getting started and it’s a fascinating transformation and that’s the fun part.
English