Boostlingo puts AI Interpreter into general availability with new language-by-language quality checks

Boostlingo says its AI Interpreter is moving from beta to general availability after new testing showed it beat a leading open-source language model by 17% in accuracy. The company is using a new system called Assure to approve each language separately, with 11 languages ready for live service and 16 still in beta.

Why it matters: - Boostlingo is trying to make AI interpreting usable for real service encounters, where quality and reliability can affect care, access, and compliance. - The company is moving away from a one-size-fits-all launch model and tying availability to language-level performance. - The approach gives language service providers a clearer way to decide where AI interpretation can safely supplement human interpreters.

What happened: - Boostlingo announced that AI Interpreter, its consecutive AI interpreting service, is moving to general availability. - Before GA, Boostlingo tested AI Interpreter against a leading open-source language model and said AI Interpreter outperformed it by 17% in accuracy. - Boostlingo said 11 languages are ready for general availability: Arabic, Cantonese, Dari, French, Haitian Creole, Korean, Mandarin, Portuguese, Russian, Spanish and Vietnamese. - Sixteen additional languages remain in beta and will move up on a rolling basis as they meet the same standard.

The details: - Boostlingo created Assure, a new evaluation framework, to judge whether each language is ready for live interpreting. - The system starts with automated evaluations against curated reference conversations. - Languages that pass then move to a human validation round using staged medical conversations meant to reflect clinical encounters. - Those recordings are scored across more than 40 sub-metrics covering accuracy, professionalism, flow and technical quality. - Boostlingo said no single score determines the outcome; automated scoring and human validation are weighed together. - During beta, AI Interpreter handled hundreds of thousands of minutes of live interpretation. - Boostlingo said that usage data helped improve call reliability, system performance and language-by-language accuracy. - A language that falls below the threshold will be re-evaluated before it can remain in or return to GA. - The same framework will be used to grade every new language going forward. - AI Interpreter launched on mobile only and is now available on Web, phone via IVR and SIP. - Partners can embed AI Interpreter through Boostlingo’s iOS and Android SDKs. - The service is now live across Boostlingo’s five regions: the US, Canada, Australia, the EU and the UK. - Language service provider admins can set policies for which languages route to AI Interpreter using the same governance model they already use for human interpretation.

Between the lines: - Boostlingo is signaling that AI interpretation is not ready to be treated as a single product with one launch status across all languages. - The language-by-language model suggests the company is prioritizing trust and clinical readiness over speed. - The medical-conversation testing focus shows Boostlingo is positioning AI Interpreter for high-stakes conversations, not just general customer support. - By building Assure in-house, Boostlingo is making quality evaluation part of the product itself instead of relying on a generic external benchmark.

What's next: - Boostlingo said it will continue expanding visibility into quality across every call. - The 16 beta languages will graduate as they clear the same evaluation standard. - Any new language added to AI Interpreter will go through the Assure framework before launch.

The bottom line: - Boostlingo is pairing AI Interpreter’s GA launch with a stricter quality gate that approves languages one by one, rather than turning the whole product on at once.