News
Research Report
The AI Transformation 100: Concrete Ideas for Fixing How We Work
Key Takeaways
- On-device AI tools. Developers can add summarization, translation and moderation directly in Chrome.
- Cost and privacy. Built-in AI reduces cloud costs and keeps user data local.
- Developer impact. Web app builders gain efficient AI features without extra infrastructure or privacy risk.
On May 26, 2026, Google introduced a suite of on-device AI capabilities built directly into Chrome, giving web developers access to summarization, translation, content moderation and accessibility tools without relying on cloud infrastructure. Thomas Steiner presented the features at Google I/O 2026.
By running AI models locally in the browser, the new APIs aim to eliminate cloud inference costs and keep sensitive user data on-device, according to Google. Partners including Drupal, Yahoo! Japan and Trip.com have already adopted the APIs in production environments.
Google also collaborated with Build Awesome (formerly Eleventy) to release a blog template that integrates all the announced AI features, giving developers a starting point for implementation.
Table of Contents
- Google's Built-In AI APIs: Feature Breakdown
- Google’s Partners Already Have APIs in Production
- Recent Google News
Google's Built-In AI APIs: Feature Breakdown
Google announced six APIs and a developer template spanning content generation, moderation, accessibility and translation.
| Capability | How It Works |
|---|---|
| Summarizer API | Generates headlines and SEO meta descriptions from article text |
| Prompt API (structured output) | Returns predictable JSON for tag generation and comment moderation |
| Prompt API (multimodal input) | Analyzes images to generate alt-text and captions |
| Writer API | Expands bullet points into full paragraphs |
| Rewriter API | Can change tone or length of text |
| Translator API | Translates drafted content into other languages on-device |
Google’s Partners Already Have APIs in Production
According to Google, many of its partners already have these APIs in production. Examples they shared include:
- Drupal uses the Summarizer API for SEO tag generation
- Yahoo! Japan uses the Prompt API for community comment moderation
- Trip.com uses AI overviews to helps shoppers navigate complex flight bookings
Google also offers a starter template that walks users through how to use these AI features.
Recent Google News
On the AI front: Gemini 3 Deep Think set new scientific reasoning benchmarks surpassing Anthropic and OpenAI, Apple agreed to use Gemini to power its revamped Siri and Google closed its $32 billion acquisition of Wiz — the largest cybersecurity deal on record.
On the agentic and commerce front, Google launched the Universal Commerce Protocol in January 2026 to address fragmented infrastructure blocking agentic AI, while Google Antigravity added Agent Skills standard support for portable AI agent capabilities.
At Google I/O 2026, the company introduced the Gemini 3.5 family, the Gemini Omni video model and the Gemini Spark autonomous agent, alongside an overhauled Search experience that has surpassed 1 billion monthly AI Mode users.
Despite this forward momentum, regulatory headwinds remain significant. A US federal judge barred Google from exclusive search deals in September 2025, and a subsequent denial of Google's request to pause those remedies triggered immediate compliance in May 2026.