Exposing Product Bias in LLM Investment Recommendation Article Swipe
YOU?
·
· 2025
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2503.08750
Large language models (LLMs), as a new generation of recommendation engines, possess powerful summarization and data analysis capabilities, surpassing traditional recommendation systems in both scope and performance. One promising application is investment recommendation. In this paper, we reveal a novel product bias in LLM investment recommendation, where LLMs exhibit systematic preferences for specific products. Such preferences can subtly influence user investment decisions, potentially leading to inflated valuations of products and financial bubbles, posing risks to both individual investors and market stability. To comprehensively study the product bias, we develop an automated pipeline to create a dataset of 567,000 samples across five asset classes (stocks, mutual funds, cryptocurrencies, savings, and portfolios). With this dataset, we present the bf first study on product bias in LLM investment recommendations. Our findings reveal that LLMs exhibit clear product preferences, such as certain stocks (e.g., `AAPL' from Apple and `MSFT' from Microsoft). Notably, this bias persists even after applying debiasing techniques. We urge AI researchers to take heed of the product bias in LLM investment recommendations and its implications, ensuring fairness and security in the digital space and market.
Related Topics
- Type
- preprint
- Language
- en
- Landing Page
- http://arxiv.org/abs/2503.08750
- https://arxiv.org/pdf/2503.08750
- OA Status
- green
- OpenAlex ID
- https://openalex.org/W4415100764
Raw OpenAlex JSON
- OpenAlex ID
-
https://openalex.org/W4415100764Canonical identifier for this work in OpenAlex
- DOI
-
https://doi.org/10.48550/arxiv.2503.08750Digital Object Identifier
- Title
-
Exposing Product Bias in LLM Investment RecommendationWork title
- Type
-
preprintOpenAlex work type
- Language
-
enPrimary language
- Publication year
-
2025Year of publication
- Publication date
-
2025-03-11Full publication date if available
- Authors
-
Yuhan Zhi, Xiaoyu Zhang, L. Wang, Shumin Jiang, Shiqing Ma, Xiaohong Guan, Chao ShenList of authors in order
- Landing page
-
https://arxiv.org/abs/2503.08750Publisher landing page
- PDF URL
-
https://arxiv.org/pdf/2503.08750Direct link to full text PDF
- Open access
-
YesWhether a free full text is available
- OA status
-
greenOpen access status per OpenAlex
- OA URL
-
https://arxiv.org/pdf/2503.08750Direct OA link when available
- Cited by
-
0Total citation count in OpenAlex
Full payload
| id | https://openalex.org/W4415100764 |
|---|---|
| doi | https://doi.org/10.48550/arxiv.2503.08750 |
| ids.doi | https://doi.org/10.48550/arxiv.2503.08750 |
| ids.openalex | https://openalex.org/W4415100764 |
| fwci | 0.0 |
| type | preprint |
| title | Exposing Product Bias in LLM Investment Recommendation |
| biblio.issue | |
| biblio.volume | |
| biblio.last_page | |
| biblio.first_page | |
| topics[0].id | https://openalex.org/T11891 |
| topics[0].field.id | https://openalex.org/fields/14 |
| topics[0].field.display_name | Business, Management and Accounting |
| topics[0].score | 0.2924000024795532 |
| topics[0].domain.id | https://openalex.org/domains/2 |
| topics[0].domain.display_name | Social Sciences |
| topics[0].subfield.id | https://openalex.org/subfields/1404 |
| topics[0].subfield.display_name | Management Information Systems |
| topics[0].display_name | Big Data and Business Intelligence |
| topics[1].id | https://openalex.org/T12276 |
| topics[1].field.id | https://openalex.org/fields/14 |
| topics[1].field.display_name | Business, Management and Accounting |
| topics[1].score | 0.2542000114917755 |
| topics[1].domain.id | https://openalex.org/domains/2 |
| topics[1].domain.display_name | Social Sciences |
| topics[1].subfield.id | https://openalex.org/subfields/1402 |
| topics[1].subfield.display_name | Accounting |
| topics[1].display_name | Risk Management in Financial Firms |
| is_xpac | False |
| apc_list | |
| apc_paid | |
| language | en |
| locations[0].id | pmh:oai:arXiv.org:2503.08750 |
| locations[0].is_oa | True |
| locations[0].source.id | https://openalex.org/S4306400194 |
| locations[0].source.issn | |
| locations[0].source.type | repository |
| locations[0].source.is_oa | True |
| locations[0].source.issn_l | |
| locations[0].source.is_core | False |
| locations[0].source.is_in_doaj | False |
| locations[0].source.display_name | arXiv (Cornell University) |
| locations[0].source.host_organization | https://openalex.org/I205783295 |
| locations[0].source.host_organization_name | Cornell University |
| locations[0].source.host_organization_lineage | https://openalex.org/I205783295 |
| locations[0].license | |
| locations[0].pdf_url | https://arxiv.org/pdf/2503.08750 |
| locations[0].version | submittedVersion |
| locations[0].raw_type | text |
| locations[0].license_id | |
| locations[0].is_accepted | False |
| locations[0].is_published | False |
| locations[0].raw_source_name | |
| locations[0].landing_page_url | http://arxiv.org/abs/2503.08750 |
| locations[1].id | doi:10.48550/arxiv.2503.08750 |
| locations[1].is_oa | True |
| locations[1].source.id | https://openalex.org/S4306400194 |
| locations[1].source.issn | |
| locations[1].source.type | repository |
| locations[1].source.is_oa | True |
| locations[1].source.issn_l | |
| locations[1].source.is_core | False |
| locations[1].source.is_in_doaj | False |
| locations[1].source.display_name | arXiv (Cornell University) |
| locations[1].source.host_organization | https://openalex.org/I205783295 |
| locations[1].source.host_organization_name | Cornell University |
| locations[1].source.host_organization_lineage | https://openalex.org/I205783295 |
| locations[1].license | cc-by |
| locations[1].pdf_url | |
| locations[1].version | |
| locations[1].raw_type | article |
| locations[1].license_id | https://openalex.org/licenses/cc-by |
| locations[1].is_accepted | False |
| locations[1].is_published | |
| locations[1].raw_source_name | |
| locations[1].landing_page_url | https://doi.org/10.48550/arxiv.2503.08750 |
| indexed_in | arxiv, datacite |
| authorships[0].author.id | https://openalex.org/A5101935015 |
| authorships[0].author.orcid | https://orcid.org/0000-0003-4977-3656 |
| authorships[0].author.display_name | Yuhan Zhi |
| authorships[0].author_position | first |
| authorships[0].raw_author_name | Zhi, Yuhan |
| authorships[0].is_corresponding | False |
| authorships[1].author.id | https://openalex.org/A5100419400 |
| authorships[1].author.orcid | https://orcid.org/0000-0002-2041-9955 |
| authorships[1].author.display_name | Xiaoyu Zhang |
| authorships[1].author_position | middle |
| authorships[1].raw_author_name | Zhang, Xiaoyu |
| authorships[1].is_corresponding | False |
| authorships[2].author.id | https://openalex.org/A5107247784 |
| authorships[2].author.orcid | https://orcid.org/0009-0005-9145-743X |
| authorships[2].author.display_name | L. Wang |
| authorships[2].author_position | middle |
| authorships[2].raw_author_name | Wang, Longtian |
| authorships[2].is_corresponding | False |
| authorships[3].author.id | https://openalex.org/A5039077071 |
| authorships[3].author.orcid | https://orcid.org/0000-0003-2421-5711 |
| authorships[3].author.display_name | Shumin Jiang |
| authorships[3].author_position | middle |
| authorships[3].raw_author_name | Jiang, Shumin |
| authorships[3].is_corresponding | False |
| authorships[4].author.id | https://openalex.org/A5101594068 |
| authorships[4].author.orcid | https://orcid.org/0000-0003-1551-8948 |
| authorships[4].author.display_name | Shiqing Ma |
| authorships[4].author_position | middle |
| authorships[4].raw_author_name | Ma, Shiqing |
| authorships[4].is_corresponding | False |
| authorships[5].author.id | https://openalex.org/A5044978005 |
| authorships[5].author.orcid | |
| authorships[5].author.display_name | Xiaohong Guan |
| authorships[5].author_position | middle |
| authorships[5].raw_author_name | Guan, Xiaohong |
| authorships[5].is_corresponding | False |
| authorships[6].author.id | https://openalex.org/A5039560181 |
| authorships[6].author.orcid | https://orcid.org/0000-0002-0619-1309 |
| authorships[6].author.display_name | Chao Shen |
| authorships[6].author_position | last |
| authorships[6].raw_author_name | Shen, Chao |
| authorships[6].is_corresponding | False |
| has_content.pdf | False |
| has_content.grobid_xml | False |
| is_paratext | False |
| open_access.is_oa | True |
| open_access.oa_url | https://arxiv.org/pdf/2503.08750 |
| open_access.oa_status | green |
| open_access.any_repository_has_fulltext | False |
| created_date | 2025-10-13T00:00:00 |
| display_name | Exposing Product Bias in LLM Investment Recommendation |
| has_fulltext | False |
| is_retracted | False |
| updated_date | 2025-11-06T06:51:31.235846 |
| primary_topic.id | https://openalex.org/T11891 |
| primary_topic.field.id | https://openalex.org/fields/14 |
| primary_topic.field.display_name | Business, Management and Accounting |
| primary_topic.score | 0.2924000024795532 |
| primary_topic.domain.id | https://openalex.org/domains/2 |
| primary_topic.domain.display_name | Social Sciences |
| primary_topic.subfield.id | https://openalex.org/subfields/1404 |
| primary_topic.subfield.display_name | Management Information Systems |
| primary_topic.display_name | Big Data and Business Intelligence |
| cited_by_count | 0 |
| locations_count | 2 |
| best_oa_location.id | pmh:oai:arXiv.org:2503.08750 |
| best_oa_location.is_oa | True |
| best_oa_location.source.id | https://openalex.org/S4306400194 |
| best_oa_location.source.issn | |
| best_oa_location.source.type | repository |
| best_oa_location.source.is_oa | True |
| best_oa_location.source.issn_l | |
| best_oa_location.source.is_core | False |
| best_oa_location.source.is_in_doaj | False |
| best_oa_location.source.display_name | arXiv (Cornell University) |
| best_oa_location.source.host_organization | https://openalex.org/I205783295 |
| best_oa_location.source.host_organization_name | Cornell University |
| best_oa_location.source.host_organization_lineage | https://openalex.org/I205783295 |
| best_oa_location.license | |
| best_oa_location.pdf_url | https://arxiv.org/pdf/2503.08750 |
| best_oa_location.version | submittedVersion |
| best_oa_location.raw_type | text |
| best_oa_location.license_id | |
| best_oa_location.is_accepted | False |
| best_oa_location.is_published | False |
| best_oa_location.raw_source_name | |
| best_oa_location.landing_page_url | http://arxiv.org/abs/2503.08750 |
| primary_location.id | pmh:oai:arXiv.org:2503.08750 |
| primary_location.is_oa | True |
| primary_location.source.id | https://openalex.org/S4306400194 |
| primary_location.source.issn | |
| primary_location.source.type | repository |
| primary_location.source.is_oa | True |
| primary_location.source.issn_l | |
| primary_location.source.is_core | False |
| primary_location.source.is_in_doaj | False |
| primary_location.source.display_name | arXiv (Cornell University) |
| primary_location.source.host_organization | https://openalex.org/I205783295 |
| primary_location.source.host_organization_name | Cornell University |
| primary_location.source.host_organization_lineage | https://openalex.org/I205783295 |
| primary_location.license | |
| primary_location.pdf_url | https://arxiv.org/pdf/2503.08750 |
| primary_location.version | submittedVersion |
| primary_location.raw_type | text |
| primary_location.license_id | |
| primary_location.is_accepted | False |
| primary_location.is_published | False |
| primary_location.raw_source_name | |
| primary_location.landing_page_url | http://arxiv.org/abs/2503.08750 |
| publication_date | 2025-03-11 |
| publication_year | 2025 |
| referenced_works_count | 0 |
| abstract_inverted_index.a | 5, 38, 94 |
| abstract_inverted_index.AI | 158 |
| abstract_inverted_index.In | 33 |
| abstract_inverted_index.To | 81 |
| abstract_inverted_index.We | 156 |
| abstract_inverted_index.an | 89 |
| abstract_inverted_index.as | 4, 136 |
| abstract_inverted_index.bf | 116 |
| abstract_inverted_index.in | 22, 42, 122, 167, 178 |
| abstract_inverted_index.is | 30 |
| abstract_inverted_index.of | 8, 67, 96, 163 |
| abstract_inverted_index.on | 119 |
| abstract_inverted_index.to | 64, 74, 92, 160 |
| abstract_inverted_index.we | 36, 87, 113 |
| abstract_inverted_index.LLM | 43, 123, 168 |
| abstract_inverted_index.One | 27 |
| abstract_inverted_index.Our | 126 |
| abstract_inverted_index.and | 14, 25, 69, 78, 108, 143, 171, 176, 182 |
| abstract_inverted_index.can | 56 |
| abstract_inverted_index.for | 51 |
| abstract_inverted_index.its | 172 |
| abstract_inverted_index.new | 6 |
| abstract_inverted_index.the | 84, 115, 164, 179 |
| abstract_inverted_index.LLMs | 47, 130 |
| abstract_inverted_index.Such | 54 |
| abstract_inverted_index.With | 110 |
| abstract_inverted_index.bias | 41, 121, 149, 166 |
| abstract_inverted_index.both | 23, 75 |
| abstract_inverted_index.data | 15 |
| abstract_inverted_index.even | 151 |
| abstract_inverted_index.five | 100 |
| abstract_inverted_index.from | 141, 145 |
| abstract_inverted_index.heed | 162 |
| abstract_inverted_index.such | 135 |
| abstract_inverted_index.take | 161 |
| abstract_inverted_index.that | 129 |
| abstract_inverted_index.this | 34, 111, 148 |
| abstract_inverted_index.urge | 157 |
| abstract_inverted_index.user | 59 |
| abstract_inverted_index.Apple | 142 |
| abstract_inverted_index.Large | 0 |
| abstract_inverted_index.after | 152 |
| abstract_inverted_index.asset | 101 |
| abstract_inverted_index.bias, | 86 |
| abstract_inverted_index.clear | 132 |
| abstract_inverted_index.first | 117 |
| abstract_inverted_index.novel | 39 |
| abstract_inverted_index.risks | 73 |
| abstract_inverted_index.scope | 24 |
| abstract_inverted_index.space | 181 |
| abstract_inverted_index.study | 83, 118 |
| abstract_inverted_index.where | 46 |
| abstract_inverted_index.(e.g., | 139 |
| abstract_inverted_index.`AAPL' | 140 |
| abstract_inverted_index.`MSFT' | 144 |
| abstract_inverted_index.across | 99 |
| abstract_inverted_index.create | 93 |
| abstract_inverted_index.funds, | 105 |
| abstract_inverted_index.market | 79 |
| abstract_inverted_index.models | 2 |
| abstract_inverted_index.mutual | 104 |
| abstract_inverted_index.paper, | 35 |
| abstract_inverted_index.posing | 72 |
| abstract_inverted_index.reveal | 37, 128 |
| abstract_inverted_index.stocks | 138 |
| abstract_inverted_index.subtly | 57 |
| abstract_inverted_index.(LLMs), | 3 |
| abstract_inverted_index.567,000 | 97 |
| abstract_inverted_index.certain | 137 |
| abstract_inverted_index.classes | 102 |
| abstract_inverted_index.dataset | 95 |
| abstract_inverted_index.develop | 88 |
| abstract_inverted_index.digital | 180 |
| abstract_inverted_index.exhibit | 48, 131 |
| abstract_inverted_index.leading | 63 |
| abstract_inverted_index.market. | 183 |
| abstract_inverted_index.possess | 11 |
| abstract_inverted_index.present | 114 |
| abstract_inverted_index.product | 40, 85, 120, 133, 165 |
| abstract_inverted_index.samples | 98 |
| abstract_inverted_index.systems | 21 |
| abstract_inverted_index.(stocks, | 103 |
| abstract_inverted_index.Notably, | 147 |
| abstract_inverted_index.analysis | 16 |
| abstract_inverted_index.applying | 153 |
| abstract_inverted_index.bubbles, | 71 |
| abstract_inverted_index.dataset, | 112 |
| abstract_inverted_index.engines, | 10 |
| abstract_inverted_index.ensuring | 174 |
| abstract_inverted_index.fairness | 175 |
| abstract_inverted_index.findings | 127 |
| abstract_inverted_index.inflated | 65 |
| abstract_inverted_index.language | 1 |
| abstract_inverted_index.persists | 150 |
| abstract_inverted_index.pipeline | 91 |
| abstract_inverted_index.powerful | 12 |
| abstract_inverted_index.products | 68 |
| abstract_inverted_index.savings, | 107 |
| abstract_inverted_index.security | 177 |
| abstract_inverted_index.specific | 52 |
| abstract_inverted_index.automated | 90 |
| abstract_inverted_index.debiasing | 154 |
| abstract_inverted_index.financial | 70 |
| abstract_inverted_index.influence | 58 |
| abstract_inverted_index.investors | 77 |
| abstract_inverted_index.products. | 53 |
| abstract_inverted_index.promising | 28 |
| abstract_inverted_index.decisions, | 61 |
| abstract_inverted_index.generation | 7 |
| abstract_inverted_index.individual | 76 |
| abstract_inverted_index.investment | 31, 44, 60, 124, 169 |
| abstract_inverted_index.stability. | 80 |
| abstract_inverted_index.surpassing | 18 |
| abstract_inverted_index.systematic | 49 |
| abstract_inverted_index.valuations | 66 |
| abstract_inverted_index.Microsoft). | 146 |
| abstract_inverted_index.application | 29 |
| abstract_inverted_index.potentially | 62 |
| abstract_inverted_index.preferences | 50, 55 |
| abstract_inverted_index.researchers | 159 |
| abstract_inverted_index.techniques. | 155 |
| abstract_inverted_index.traditional | 19 |
| abstract_inverted_index.performance. | 26 |
| abstract_inverted_index.portfolios). | 109 |
| abstract_inverted_index.preferences, | 134 |
| abstract_inverted_index.capabilities, | 17 |
| abstract_inverted_index.implications, | 173 |
| abstract_inverted_index.summarization | 13 |
| abstract_inverted_index.recommendation | 9, 20 |
| abstract_inverted_index.comprehensively | 82 |
| abstract_inverted_index.recommendation, | 45 |
| abstract_inverted_index.recommendation. | 32 |
| abstract_inverted_index.recommendations | 170 |
| abstract_inverted_index.recommendations. | 125 |
| abstract_inverted_index.cryptocurrencies, | 106 |
| cited_by_percentile_year | |
| countries_distinct_count | 0 |
| institutions_distinct_count | 7 |
| citation_normalized_percentile |