Two numbers from autumn 2026 matter to every CIO. Google's Gemini 4 Argon can produce up to one million tokens in a single response — enough for a whole multi-file codebase. And in an early test, Anthropic's Claude Opus 5.5 completed a 680,000-line code migration in under a day, work that normally takes an engineering team weeks.
Old systems resist modernization because nobody fully understands them: documentation is stale and the original developers are gone. A model that can hold the entire system in view — code, database schemas, configuration and tickets — turns the first, most expensive step, simply understanding what exists, into a matter of days.
People still decide the architecture, write the tests and confirm the business logic. The teams getting results pair AI speed with strong automated testing and staged cutovers.
What it means for your business
- Re-estimate stalled modernization projects; discovery and translation are now much cheaper.
- Build automated tests first — they make AI-generated code safe to trust.
- Keep architecture and sign-off with experienced engineers.
How MTH can help: Our Infrastructure Modernization and Migration services use AI-assisted discovery and translation, with milestone-based acceptance.
This briefing is original commentary by MTH based on public reporting; figures are as reported by the sources named above and may be revised over time.
