# The AI Brand Misinformation Report

Over 1,700 brands reveal how often AI chatbots produce false facts about them.

Source: Searchable first-party data. 32,556 facts stated by AI engines about 1,704 brands, each graded correct or incorrect by an employee of the organisation it describes. Data through 29 July 2026.

## Key findings

- Brands flagged 9.2% of AI-stated facts about them as false (3,005 of 32,556).
- AI is most likely to be wrong about pricing (13.5%), funding and financials (10.9%), and the software and tools a company integrates with (10.6%).
- Of brands that graded at least 20 facts, 75% found at least one fact to be false.
- AI companies see the most false facts about them, at 18.5%.
- New Zealand (15.7%), Germany (15.0%) and Canada (14.3%) are the worst-affected markets.
- Small businesses (1–50 staff) see false facts at 9.4%, against 7.4% for mid-size and 6.3% for large firms.

## 1. How often do brands find incorrect AI-generated facts about them?

Brands flagged 9.2% of the facts AI engines stated about them as incorrect: 3,005 false facts out of 32,556 reviewed.

AI answers name only a handful of brands and state only a handful of facts about each, so a roughly one-in-eleven error rate is materially different from one bad result on a page of ten blue links.

## 2. What type of fact does AI get wrong most often about brands?

False-fact rate by type of claim, with reviewed-fact volume:

| Rank | Type of fact | % false | Reviewed facts |
|---|---|---|---|
| 1 | Pricing | 13.5% | 2,907 |
| 2 | Funding & financials | 10.9% | 488 |
| 3 | Integrations & tech stack | 10.6% | 2,891 |
| 4 | Partnerships & clients | 9.9% | 3,822 |
| 5 | Reviews & ratings | 8.8% | 758 |
| 6 | Products & services | 8.6% | 20,173 |
| 7 | Mergers & acquisitions | 7.2% | 138 |
| 8 | History & founding | 6.9% | 173 |
| 9 | Location & coverage | 4.8% | 827 |
| 10 | Awards & certifications | 3.4% | 379 |

The facts that change most often are the ones AI is most likely to be out of date on, and they are also the facts that influence whether someone buys.

## 3. What percentage of brands have found false facts about them on AI chatbots?

Among brands that reviewed at least 20 facts about them, 75% found at least one fact to be false.

Brands that spot-check only two or three facts often find nothing wrong, because they have not looked hard enough for the error rate to surface.

## 4. Which sectors are worst-affected by false AI facts?

| # | Sector | % false |
|---|---|---|
| 1 | AI / Machine Learning | 18.5% |
| 2 | Professional Services / Consulting | 10.8% |
| 3 | SaaS / Software | 9.5% |
| 4 | Travel / Tourism | 9.2% |
| 5 | Health / Wellness / Medical | 8.4% |
| 6 | Food & Beverage / Hospitality | 8.2% |
| 7 | Marketing / Advertising | 8.1% |
| 8 | Financial Services / Fintech | 8.0% |
| 9 | E-commerce / Retail | 7.6% |
| 10 | Media / Entertainment | 7.6% |

AI companies tend to be younger, with less settled authoritative information online for AI systems to draw on.

## 5. The countries where companies are worst-affected by false AI facts

| Rank | Market | % false | Reviewed facts |
|---|---|---|---|
| 1 | New Zealand | 15.7% | 460 |
| 2 | Germany | 15.0% | 868 |
| 3 | Canada | 14.3% | 704 |
| 4 | Spain | 14.3% | 273 |
| 5 | Netherlands | 13.6% | 279 |
| 6 | Australia | 11.7% | 1,987 |
| 7 | UAE | 10.3% | 689 |
| 8 | South Africa | 10.0% | 629 |
| 9 | Singapore | 9.8% | 338 |
| 10 | Sweden | 9.3% | 572 |
| 11 | Italy | 9.1% | 405 |
| 12 | United States | 8.6% | 12,131 |
| 13 | United Kingdom | 8.2% | 6,915 |
| 14 | Finland | 7.7% | 1,160 |

Markets need at least 250 reviewed facts to appear in the ranking.

This is not simply an English-language effect: New Zealand and Canada are English-speaking yet sit at the opposite end of the table from the US and UK. A more likely explanation is how much of the internet's information trail is US and UK-centric, with language a compounding factor — English has historically dominated LLM training data, with GPT-3's corpus over 92% English.

## 6. The smaller the business, the more AI gets wrong

| Company size | % false | Brands | Reviewed facts |
|---|---|---|---|
| Small (1–50 staff) | 9.4% | 205 | 12,946 |
| Mid-size (51–250) | 7.4% | 35 | 1,791 |
| Large (250+) | 6.3% | 20 | 1,503 |

Small businesses see false facts about 50% more often than large firms. 77.6% of small businesses found at least one false fact about themselves, the highest of any size band.

Where AI trips up on small firms: tools and tech stack (12.8%), pricing (11.8%), partnerships (10.0%).

A smaller business often has a thinner digital footprint, so AI has less to work from and fills the gaps with guesses.

## 7. US brands come off better than most, except on price

- 8.6% of US brands' facts were flagged false, against 9.6% across the rest of the world.
- 69.4% of US brands that checked 20 or more facts found at least one that was false, against 78.6% elsewhere.
- Where AI trips up on US brands: pricing (16.1%), tools and tech stack (9.4%), partnerships (9.1%).

US pricing errors run at 16.1%, well above the 13.5% global figure and nearly double these brands' own overall rate.

## 8. How brands combat AI hallucinations

1. Get your Organization schema and sameAs links right.
2. Tell the same story everywhere — name, location, founding date and description consistent across site, socials, directories and press.
3. Claim and populate a Wikidata entry.
4. Earn credible third-party coverage.
5. Review and correct the facts in-platform, and re-check periodically.

## Methodology

- Searchable surfaces the facts AI engines state about a brand and lets the brand review each one and mark it correct or incorrect. 32,556 reviewed facts across 1,704 brands, to 29 July 2026.
- ChatGPT accounts for 20,359 reviewed facts (8.1% false), Perplexity 6,676 (11.5%) and Google AI Overviews 4,881 (11.4%). Copilot, Gemini, Google AI Mode, Claude, Grok and DeepSeek make up the remainder.
- Each fact is manually graded by an employee of the organisation it describes. No automated counting.
- "False" covers three groups: outdated but once-true details, mistaken identity where AI confuses one brand with another, and genuine fabrications. Hallucination is reserved for the fabricated group only.
- Sections 3, 6 and 7 count brands that reviewed 20 facts or more. Fact-level error rates hold at roughly 9% at every cut-off.
- Market figures cover markets with at least 250 reviewed facts. All markets remain in every whole-dataset figure.

You are free to reuse and reinterpret this research, with linked credit to https://www.searchable.com/data/ai-brand-misinformation-report