# Google now runs two different searches — and you optimize for one

> AI Overview and AI Mode run on different engines: AIO leans on the classic results, AIM is vector semantic search. The data (600+ parameters) and the two distinct strategies.

_Source: https://seomatrix.ai/blog/aio-vs-aim/ · Updated: 2026-07-06_

---

## Two searches behind one button

The biggest architectural discovery about Google in 2026: **AI Overview (AIO)** and **AI Mode (AIM)** aren't one "AI search" in two skins — they're **two different systems on different retrieval engines**. The hypothesis (Ashmanov & Partners, empirics across 600+ parameters): AIO assembles its answer from the classic results, AIM searches a vector semantic index, the way ChatGPT does.

Practically: "one template for everything" doesn't work. Optimizing only for the classic results covers AIO — and leaves you nearly nonexistent in AIM.

## The data: how they differ

| Parameter | AI Overview (AIO) | AI Mode (AIM) |
|---|---|---|
| Answer length | short (1–2 paragraphs) | **~4× longer** |
| Sources from the original query | **~60%** | **~15%** |
| + from extra sub-queries | 10–15% | ~30% |
| Top-3 sources | ~40% of slots | flatter distribution |
| Commercial pages among sources | 45.6% | 29.2% |
| Source languages | as in the results (monolingual) | multilingual |
| Ambiguous words | one meaning | all meanings |
| Correlation with organic | stronger | weaker |

The most telling piece is the **histogram shape**. In AIO the source-by-position distribution looks like classic search: dense on the top-3, exponential decay. In AIM it's flat — exactly what you'd expect from an embedding-index search, where SERP position means nothing and semantic similarity decides.

> **A hypothesis — but a working one** — US v. Google records and the leaks show no explicit engine split. But the empirics (histograms, languages, commercial share, ambiguous words) fit the known architecture: FastSearch + MUVERA as a separate vector retrieval system. You can build the strategy on it today.

## The AIO strategy: the classics still work

- **Top-10 still supplies ~60% of sources** — rankings remain a lever.
- **[Query fan-out](/blog/query-fan-out/)** gives pages outside the top-3 a shot — via sub-queries.
- **Freshness wins**: recently updated pages systematically take the slots.
- The commercial share is high (45.6%) — commercial pages are allowed into AIO.

## The AIM strategy: position barely matters

- **Top-10 supplies only ~15% of sources** — ranking barely helps.
- You need **vector similarity, not lexical match**: [self-contained atomic blocks of 40–60 words](/tools/ai-view/), a number at the start of the passage, structured data.
- **Multilingual**: AIM cites foreign sources — your content can surface in other locales (and competitors arrive from other languages).
- The answer is 4× longer → **more citation slots**, but each slot is won by a separate sub-intent (see [where engines get their answers](/blog/ai-citation-sources/)).

## What to do

1. **Split the KPIs.** Presence in AIO and in AIM are two different results; measure them separately, like rankings.
2. **For AIO** — keep the classics: rankings, freshness, sub-query fan coverage.
3. **For AIM** — atomize: every sub-intent = a self-contained block, extractable without context.
4. **Check both surfaces.** Our AI-visibility probe measures citation across 4 engines, and geo-metrics separately counts [your AI-Overview exposure per keyword](/tools/ai-overview-checker/).

> **Check it in 2 minutes** — Run your domain through the [AI-visibility check](https://seomatrix.ai/ai-visibility/) — see where you're cited (and whether at all), for which queries, and who's cited instead.

## Sources

- **GEO-HowTo 2026 — Kurdyukova–Volovich talk (Ashmanov & Partners)** — the two-engine hypothesis, a 600+-parameter methodology: source histograms, original-query shares (~60% vs ~15%), commercial pages (45.6% vs 29.2%), language behavior.
- **GEO-HowTo 2026 — Alexander Runov's talk (MGCom)** — Google Search 2.0 (2026-05-19): "AI became the foundation of the search interface."
- **US v. Google case materials (2025)** — FastSearch as a separate retrieval system (embeddings, dot-product) — the architectural basis of the hypothesis.

## FAQ

### How is AI Overview different from AI Mode?

AI Overview is a short summary above the classic results, assembled mostly from them (~60% of sources from the original query). AI Mode is a separate conversational surface with an answer ~4× longer, assembled by vector semantic search (~15% of sources from the original query, the rest via sub-queries and semantic similarity).

### Is it true that AIO and AIM run on different engines?

It's a hypothesis by Ashmanov & Partners, empirically grounded on 600+ parameters: the shape of source histograms, language behavior, the share of commercial pages (45.6% in AIO vs 29.2% in AIM), and the handling of ambiguous words. Court records and leaks show no explicit split, but the hypothesis fits the FastSearch/MUVERA architecture as a separate retrieval system.

### How do you optimize for each?

For AIO classic SEO works: top-10 supplies ~60% of sources, plus freshness and query fan-out. For AIM position barely matters — you need self-contained atomic blocks with numbers up front, structured data and sub-intent coverage: the engine searches by vector similarity, not lexical match or rank.