ISCInternational Solutions CentralGoing global, both directions

Operations

Launch in New Languages Faster With AI Translation Tools

By Jason Kumpf, International Advisor · November 4, 2026 · 5 min read

Launch in New Languages Faster With AI Translation Tools

Google Translate, DeepL and Claude have made high-quality translation fast and accessible. Companies are using them to reach international customers in their own language sooner.

Launching in a new language used to take months of translation, review and rework. Today, AI translation tools from Google, DeepL and Anthropic let companies produce strong first drafts of websites, product screens and support content in days.

The result is a faster path to international customers. Teams can test a new market with real localized content, learn quickly and invest further where the response is strongest. Here is how the tools have advanced and how companies are putting them to work.

Customers buy in their own language

The business case for localization is well established. In a 2020 study of 8,709 consumers across 29 countries, CSA Research found that 76% prefer to buy products with information in their own language. Forty percent said they will never buy from websites in other languages.

The same study found that 75% of consumers are more likely to buy from a brand again if customer care is offered in their language. In Germany, 57% of respondents said they buy only from local-language websites, the highest share of any country surveyed.

In other words, language is a growth lever. Every new language a company supports widens the audience that can discover, evaluate and adopt its product.

Google Translate gets a Gemini upgrade

Google has expanded Translate in both reach and quality. In June 2024, it added 110 new languages using its PaLM 2 model, bringing the total to more than 240. The additions cover more than 614 million speakers and include Cantonese, and roughly a quarter of the new languages are African.

In August 2025, Google introduced live, back-and-forth conversation translation in the Translate app across more than 70 languages, first for users in the US, India and Mexico. The feature detects pauses and accents and switches languages automatically.

In December 2025, Google brought Gemini's translation capabilities to Search and the Translate app. Google said the upgrade delivers state-of-the-art text quality on the WMT25 benchmark and handles idioms, slang and local expressions with more natural meaning. A beta feature now delivers real-time speech translation through any pair of headphones.

For global teams, these tools make everyday conversations with prospects, partners and colleagues easier in dozens of languages.

DeepL builds for business translation

DeepL, founded in Germany in 2017, has focused on translation for companies. In June 2025, it reported more than 200,000 business customers across 228 markets and added Vietnamese and Hebrew to DeepL Translator, with early access to Thai through its API.

In November 2025, DeepL announced support for more than 100 languages, including about 70 added in beta for enterprise customers. The release brought all 24 official EU languages to the platform, plus Hindi, Bengali, Tagalog, Swahili and others.

Two other launches matter for localization teams:

  • Customization Hub. A single place for glossaries, style rules and translation memories, applied automatically so that product names and preferred terms stay consistent.
  • DeepL Agent. An AI agent, now generally available, that plans and completes multi-step tasks under human supervision and connects to everyday tools such as email, calendars and CRM systems.

Claude helps adapt tone and context

Translation is only part of localization. Marketing copy, onboarding emails and help articles also need the right tone, examples and cultural references. Large language models such as Anthropic's Claude are well suited to that adaptation work.

Anthropic's published benchmarks show strong cross-language performance. On a professionally translated version of the MMLU test, Claude Sonnet 4.5 scored between 96.7% and 98.2% of its English results in Spanish, Brazilian Portuguese, French, German, Japanese, Korean and Hindi.

Anthropic's guidance for multilingual work is practical and simple:

  • State the source and target languages clearly.
  • Ask for idiomatic writing as a native speaker would phrase it.
  • Submit text in its native script rather than transliteration.
  • Consider cultural and regional context, not just word choice.

Teams often pair a translation engine with a model like Claude: one produces an accurate draft, and the other refines headlines, calls to action and examples for the local audience.

Where AI localization delivers quick wins

Not every piece of content needs the same treatment. Companies see the fastest results when they start with the assets that customers touch most often:

  • Website and landing pages. The homepage, product pages and sign-up flow shape first impressions in every market.
  • Product interface text. Menus, buttons and in-app messages are short, repetitive and well suited to glossary-driven AI translation.
  • Help center articles. Localized self-service content lets new customers find answers on their own, in their own language.
  • Onboarding and nurture emails. A welcome series in the customer's language builds trust in the first weeks of a relationship.
  • Sales materials and video subtitles. Localized decks, one-pagers and captions help local partners and sales teams tell the story well.

Starting with these high-visibility assets gives teams a quick read on demand. Longer documents, such as detailed technical guides, can follow once a market shows traction.

A faster localization workflow

Companies moving quickly into new languages tend to follow a repeatable process:

  1. Build a glossary first. List product names, feature terms and phrases that should never change, and load them into your tools.
  2. Translate with AI. Use a translation engine for the first draft of web pages, app strings and help content.
  3. Adapt for the market. Use a language model to adjust tone, swap in local examples and rewrite headlines that do not travel well.
  4. Review with native speakers. A local colleague, partner or freelance reviewer checks the most visible pages and the brand voice.
  5. Launch, measure and expand. Track sign-ups and engagement by language, then extend localization to the markets that respond best.

This approach keeps people focused on judgment and brand quality while AI handles volume. It also makes it practical to test several languages at once instead of one at a time.

From one language to many

AI translation has turned localization from a large project into a series of quick, manageable steps. A company can now publish a credible local version of its website, start conversations with international prospects and support new customers in their own language far sooner than before.

A good place to start is your top three international traffic sources. Localize your homepage, main product page and onboarding emails for those languages, ask a native speaker to review them and watch how visitors respond. The data will show you where to grow next.

Sources

  1. CSA Research, "Survey of 8,709 Consumers in 29 Countries Finds that 76% Prefer Purchasing Products with Information in their Own Language," July 7, 2020.
  2. SiliconANGLE, "Google uses AI to bring 110 new languages to Translate," June 27, 2024.
  3. Google, The Keyword, "New AI-powered live translation and language learning tools in Google Translate," August 26, 2025.
  4. Google, The Keyword, "Bringing state-of-the-art Gemini translation capabilities to Google Translate," December 12, 2025.
  5. DeepL, "DeepL accelerates global business connectivity and document translation with new languages," June 24, 2025.
  6. DeepL, "DeepL Unveils Next-Gen Language AI tools and Game-Changing Agentic Productivity Solution," November 5, 2025.
  7. Anthropic, Claude Platform Docs, "Multilingual support," accessed October 2026.

More news from International Solutions Central