Parse
IndexPricing
Sign inCheck your brand
  1. Research
Research
How often does AI recommend against a brand?

How often does AI recommend against a brand?

Published Jul 16, 2026Updated Oct 12, 2026

On this page
  • The finding
  • How we measured
  • Sources
  • More like this
  • The finding
  • How we measured
  • Sources
  • More like this

Dimitry ApollonskyFounder, Parse

DataWhat reviewed means

Not being recommended is different from being explicitly rejected. We wanted to see how often AI explicitly advises a buyer not to choose a brand.

We examined 1,290,741 reviewed brand statements across 314,909 answers from four AI engines, collected from October 2025 through July 2026. We counted explicit rejections separately from other kinds of brand statements.

AI rarely said a brand was unsuitable for the buyer’s stated need. We found 5,403 explicit rejections in 1.29 million reviewed statements about brands.

AI recommends against a brand in only 0.42% of reviewed statements

A reviewed statement is one brand described in one sentence of an AI answer, using cleaned brand names so that aliases count as the same brand. An explicit rejection is a statement that says the brand is not recommended for the stated need.

The Parse index contains 5,403 explicit rejections across 1,290,741 reviewed statements. Explicit rejection is too rare to use as a general measure of negative brand language.

Negative claims and explicit rejections

  • Negative claim6.5%
  • Explicit rejection0.42%
  • 0%2%4%6%8%
A negative claim records criticism or a limitation. An explicit rejection says the brand is not recommended for the stated need.

0.42%
of reviewed statements recommend against the brand
5,403 of 1,290,741

Negative claims are much more common than explicit rejection

A negative claim is a criticism or limitation recorded in the same reviewed statement. The Parse index contains 84,531 statements with a negative claim, or 6.5%. It contains 5,403 explicit rejections, or 0.42%.

BrightEdge reports negative sentiment by engine. This study measures the narrower decision that a brand is not recommended for a stated need. The two measures should not be combined.

statements with a negative claim
84,531

6.5%

explicit rejections
5,403

0.42%

Takeaway

A negative statement is not an explicit recommendation against the brand. Track the two outcomes separately.

Only 1.2% of answers contain an explicit rejection

The explicit rejections appear in 3,816 of 314,909 answers. They appear somewhere in 2,308 of 16,769 prompts, or 13.8%, because prompts were run more than once.

A prompt can produce an explicit rejection on one run and omit it on another. The answer rate measures what a user saw in one response.

answers with an explicit rejection
1.2%

3,816 of 314,909

prompts with any explicit rejection
13.8%

2,308 of 16,769

ChatGPT Search has the highest rejection rate

ChatGPT Search has a 0.71% rejection rate. ChatGPT is at 0.43%, Google AI Mode at 0.26%, and Google AI Overviews at 0.12%.

These rates count individual statements, and each engine produced a different mix of statements about brands. One average across all engines would hide the engine difference.

Rejection rate by engine

  • ChatGPT Search0.71%
  • ChatGPT0.43%
  • Google AI Mode0.26%
  • Google AI Overviews0.12%
  • 0%0.2%0.4%0.6%0.8%
Explicit rejections as a share of reviewed statements.

The gap between the two current engines holds on the same 15,996 prompts

We restricted ChatGPT Search and Google AI Mode to the same 15,996 organic prompts. ChatGPT Search has a 0.71% rejection rate across 495,970 reviewed statements. Google AI Mode has a 0.27% rate across 487,876 statements.

Asking both engines the same prompts rules out different questions as the full explanation. They can still name different brands and produce different numbers of statements.

ChatGPT Search
0.71%

3,522 of 495,970 statements

Google AI Mode
0.27%

1,291 of 487,876 statements

organic prompts measured on both engines
15,996

Risk warnings explain 17.9% of explicit rejections

No reason was recorded for 44.1% of explicit rejections. Risk warnings account for 17.9%, inferior comparisons 15.6%, conditional decisions 15.1%, fallbacks 6.8%, and budget-only decisions 0.31%.

A reason can tell a marketer whether the problem is safety, fit, comparison, or price. The group with no recorded reason needs a look at the actual sentences.

Reason mix for explicit rejections

  • No reason recorded44.1%
  • Risk warning17.9%
  • Inferior comparison15.6%
  • Conditional15.1%
  • Fallback6.8%
  • Budget only0.31%
  • 0%20%40%60%
Share of 5,403 explicit rejections.

Takeaway

The reason determines the response: correct bad facts, address accurate limitations, and read the cases with no recorded reason.

Workflow topics have the lowest rejection rate of the labeled topics

General topics have a 0.47% rejection rate. Pricing and contract topics are at 0.28%, feature requirements at 0.26%, and workflow topics at 0.09%.

A topic type is the repeated purpose behind a prompt. The result shows where an explicit rejection appears, not which topic type has more demand.

Rejection rate by topic type

  • General0.47%
  • Pricing and contract0.28%
  • Feature requirement0.26%
  • Workflow0.09%
  • 0%0.2%0.4%0.6%
Only labeled topic types with at least 20 explicit rejections are shown.

Blockchain and food brands have the highest large-group rates

Blockchain and Cryptocurrency has a 1.1% rejection rate. Food and Beverage is at 1.0%, Privacy and Security at 0.92%, and Financial Services at 0.80%.

Sales and Marketing is at 0.19%. Compare industry groups only after checking the sample threshold and the brands included in each group.

Highest and lowest industry-group rates that met the threshold

  • Blockchain and Cryptocurrency1.1%
  • Food and Beverage1.0%
  • Privacy and Security0.92%
  • Financial Services0.80%
  • Media and Entertainment0.23%
  • Manufacturing0.22%
  • Professional Services0.20%
  • Sales and Marketing0.19%
  • 0%0.5%1.0%1.5%
Explicit rejections as a share of reviewed statements.

Datadog and Jira Software have the most explicit rejections

Datadog has 74 explicit rejections across 4,272 reviewed statements. Jira Software has 69 across 2,800. ClickUp has 42. HelloFresh and PayPal each have 38.

This table ranks volume, not risk. Brands with more statements about them have more chances to appear in an explicit rejection.

Brands by number of explicit rejections

Explicit rejections · Top 10 of 20

  • Datadog74
  • Jira Software69
  • ClickUp42
  • HelloFresh38
  • PayPal38
  • Splunk36
  • Salesforce34
  • Ethereum34
  • BetterHelp33
  • Trupanion27
  • 020406080
Explore the data (20 rows)
Datadog744,2721.73
Jira Software692,8002.46
ClickUp424,2840.98
HelloFresh387155.31
PayPal386236.1
Splunk361,1823.05
Salesforce342,1861.56
Ethereum341,0983.1
BetterHelp336105.41
Trupanion275075.33
Tether274066.65
DraftKings231,7071.35
FanDuel231,2701.81
Talkspace227952.77
Notion214,1130.51
GoDaddy2113116.03
Asana193,1870.6
Atlassian191,0911.74
Solana197292.61
ADP186792.65
Cleaned brand names with at least 10 explicit rejections.

Blur and OpenSea have the highest rates among brands with enough data

Blur has an 8.8% rejection rate across 204 reviewed statements. OpenSea is at 7.2% across 249. Tether is at 6.7%, PayPal at 6.1%, and Healthy Paws Pet Insurance at 5.7%.

The rate table requires at least 200 reviewed statements and 10 explicit rejections. The threshold removes small samples from the leaderboard.

Brands by rejection rate

Rejection rate (%) · Top 10 of 20

  • Blur8.82%
  • OpenSea7.23%
  • Tether6.65%
  • PayPal6.1%
  • Healthy Paws Pet Insurance5.69%
  • Bitcoin5.47%
  • BetterHelp5.41%
  • Factor5.36%
  • Trupanion5.33%
  • HelloFresh5.31%
  • 0%5%10%
Explore the data (20 rows)
Blur8.8218204
OpenSea7.2318249
Tether6.6527406
PayPal6.138623
Healthy Paws Pet Insurance5.6914246
Bitcoin5.4718329
BetterHelp5.4133610
Factor5.3617317
Trupanion5.3327507
HelloFresh5.3138715
Jenkins4.8511227
ASPCA4.2611258
Green Chef4.2217403
Novo4.1914334
Embrace Pet Insurance3.6111305
Robinhood3.5318510
Elasticsearch3.515429
Vanguard3.3614417
EveryPlate3.2514431
Ethereum3.1341,098
At least 200 reviewed statements and 10 explicit rejections.

We left out statements that were unclear or unverified

The measured set contained 1,864,552 candidate statements and five duplicate records. We counted each record once.

We excluded 162,188 statements from prompts outside the organic prompt set, 272,043 statements we could not match to a known brand, and 155,946 statements that had not passed source review. These groups can overlap. That leaves 1,290,741 reviewed statements as the base for every rate here.

candidate statements
1,864,552
outside the organic prompt set
162,188
with no matching known brand
272,043
did not pass source review
155,946
reviewed statements included
1,290,741

What marketers should do

The Parse index contains explicit rejections for 2,557 distinct brands across 2,308 prompts. A general sentiment score cannot show which topic produced the decision.

Audit the answer sentence, topic, and reason when a brand is explicitly rejected. Correct a factual error when the answer is wrong. Address the product limitation when the answer is accurate. Then rerun the same prompt across the same engines.

distinct brands explicitly rejected
2,557
prompts with any explicit rejection
2,308

Takeaway

Criticism and explicit rejection are different outcomes. Measure both.
  • Being cited is not enough: how to monitor sentiment in AI answers

How we measured

We analyzed 1,290,741 reviewed statements across 314,909 AI answers, 16,769 organic prompts, and 139,399 distinct brands on ChatGPT, ChatGPT Search, Google AI Overviews, and Google AI Mode from October 19, 2025 through July 9, 2026.

rejection rate
0.42%
reviewed AI statements about brands
1.3M
explicit rejections
5,403
distinct brands explicitly rejected
2,557

Get the data

Dataset CSVThe metrics behind every figure in this report.

Sources

  1. BrightEdge: Google AI Overviews are 44% more likely to criticize brands than ChatGPT (opens in a new tab) · accessed July 16, 2026
  2. Semrush: 2026 AI search methodology (opens in a new tab) · accessed July 16, 2026
  3. EMNLP 2025: Bias beware in product recommendations (opens in a new tab) · accessed July 16, 2026
  4. Consumer product recommendation bias in large language models (opens in a new tab) · accessed July 16, 2026

More like this

How often does AI recommend a brand with reservations?
AI attached a reservation to a ranked brand in about one in ten appearances.
How often does a mentioned brand rank first?
Only about one in eight brands mentioned in an AI answer was its actual pick.
What words does AI use to describe brands?
Most brand descriptions were positive, and much of the praise was shared across brands.
How AI picks a winner in head-to-head comparisons
The winner often changed with the criterion being compared. A brand could win on one need and lose on another.
Brands AI recommends on just one engine
Most brands recommended in our matched questions appeared on only one of ChatGPT and Google AI Overviews.
Does AI criticize its top recommendation?
AI occasionally paired its first recommendation with a specific negative claim about the same brand.
Where does AI make a clear recommendation?
How often AI expressed reservations varied by category. Reservations appeared in 1.6% of reviewed ranked-brand appearances in warehouse robotics, compared with 22.6% in Ethereum DeFi tokens.
How often does AI call a brand an alternative?
AI explicitly called a ranked brand an alternative in about one in twenty-four appearances.
What reasons does AI give for its first recommendation?
AI cited reputation more often for first choices than lower-ranked brands. Price was the first choice's most common reservation.
Is AI getting more confident over time?
AI used less uncertain wording over time, but ChatGPT attached more conditions to its brand recommendations.
Which brands does AI tell you to avoid?
AI named brands to avoid much more often when explicitly asked. Unprompted warnings were less common.

Find the topics where AI recommends against your brand.

Parse

Index

  • Brands
  • Markets
  • Topics
  • How we measure

For agents

  • Docs
  • Connect
  • Workflows
  • llms.txt

Parse

  • Pricing
  • Research
  • Blog
  • Compare

Company

  • About
  • Contact
  • Corrections

© 2026 Parse, Inc. All rights reserved.

Privacy PolicyTerms of Service