Sony rolls out a standard way to measure bias in how AI describes what it 'sees'
(2025/11/05)
- Reference: 1762377112
- News link: https://www.theregister.co.uk/2025/11/05/sony_ai_vision_model_benchmark/
- Source link:
AI models are filled to the brim with bias, whether that's showing you a certain race of person when you ask for a pic of a criminal or assuming that a woman can't possibly be involved in a particular career when you ask for a firefighter. To deal with these issues, Sony AI has released a new dataset for testing the fairness of computer vision models, one that its makers claim was compiled in a fair and ethical way.
The [1]Fair Human-Centric Image Benchmark (FHIBE, or "Fee-bee") "is the first publicly available, consensually collected, and globally diverse fairness evaluation dataset for a wide variety of human-centric computer vision tasks," [2]according to Sony AI .
"A common misconception is that because computer vision is rooted in data and algorithms, it's a completely objective reflection of people," explains Alice Xiang, global head of AI Governance at Sony Group and lead research scientist for Sony AI, in [3]a video about the benchmark release. "But that's not the case. Computer vision can warp things depending on the biases reflected in its training data."
[4]
AI models, one way or another, [5]present bias , which perhaps with work can be minimized. Computer vision models may exhibit bias by inappropriately classifying people with specific physical characteristics in terms of occupation or some other label. They may categorize female doctors as nurses, for example.
[6]
[7]
According to Xiang, there have been instances in China where facial recognition systems on mobile phones have mistakenly allowed the device owner's family members to unlock the phone and make payments, an error [8]she speculates could come from a lack of images of Asian people in model training data or from undetected bias in the model.
"Much of the [9]controversy around facial recognition technologies has centered on their potential to be biased, leading to wrongful arrests, security breaches, and other harm," Xiang told The Register in an email. "GenAI models have also been [10]shown to be biased , reflecting harmful stereotypes."
[11]
There are a variety of [12]benchmark testing datasets for assessing the fairness of computer vision models. Meta, which in 2023 [13]disbanded its Responsible AI division but still likes to talk about [14]AI safety , has a computer vision benchmark called [15]FACET (FAirness in Computer Vision EvaluaTion).
But Xiang says the images that went into prior data sets were not gathered with the consent of those depicted – claims echoed in various AI copyright lawsuits now winding their way through the courts.
"The vast majority of computer vision benchmark datasets were collected without consent, and some were collected with consent but provide little information about the consent process, are not globally diverse, and are not suitable for a wide variety of computer vision tasks," she said.
[16]Oak Ridge lab bags $125M to bolt quantum onto supercomputers
[17]Attackers abuse Gemini AI to develop 'Thinking Robot' malware and data processing agent for spying purposes
[18]Amazon complains that Perplexity's agentic shopping bot is a terrible customer
[19]UK judge delivers a 'damp squib' in Getty AI training case, no clear precedent set
As detailed in [20]research published in Nature , the majority of the 27 evaluated computer vision datasets "were scraped from Internet platforms or derived from scraped datasets. Seven well-known datasets were revoked by their authors and are no longer publicly available."
The Sony AI authors also say that their image annotations (bounding boxes, segmentation masks, and camera settings, etc.) provide more detail, thus making FHIBE more useful to model developers.
[21]
FHIBE consists of "10,318 consensually-sourced images of 1,981 unique subjects, each with extensive and precise annotations." The images were gathered from more than 81 countries/regions.
The Sony AI researchers say they've used FHIBE to confirm that some vision models make less accurate predictions for those using "She/Her/Hers" pronouns due to the variability of hairstyles in related images. They also found that a model, when asked a neutral question about a person's occupation, would sometimes respond in a way that reinforced stereotypes by associating certain demographic groups with criminal activities.
The goal of the project, the AI group says, is to encourage more ethical and responsible practices for data gathering, use, and management.
Presently, in the US, there's not much federal government support for such sentiment. The Trump administration's "America's AI Action Plan"
According to Xiang, Sony business units have employed FHIBE in fairness assessments as part of their broader AI ethics review processes, in compliance with Sony Group AI Ethics Guidelines.
"FHIBE not only enables developers to audit their AI systems for bias but also shows that it is feasible to implement best practices in ethical data collection, particularly around consent and compensation for data rightsholders," said Xiang in [23]a social media post .
"At a time when data nihilism is increasingly common in AI, FHIBE strives to raise the standards for ethical data collection across the industry."
Asked to elaborate on this, Xiang said, "By 'data nihilism,' I mean the industry belief that data for AI development cannot be sourced consensually and with compensation, and that if we want cutting-edge AI technologies, we need to give up these data rights.
"FHIBE doesn't fully solve this problem since there's still the scalability issue (FHIBE is a small evaluation dataset, not a large training dataset), but one of our goals was to inspire the R&D community and industry to invest more care and funding into ethical data curation. This is an incredibly important problem – arguably one of the biggest problems in AI now – but far less attention is paid to innovation on the data layer compared to the algorithmic layer." ®
Get our [24]Tech Resources
[1] http://fairnessbenchmark.ai.sony
[2] https://ai.sony/articles/Groundbreaking-Fairness-Evaluation-Dataset-From-Sony%20AI%20/
[3] https://youtu.be/_V4SxY5Kqfk?si=NeuPdDwWYML4QLdJ&t=13
[4] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=2&c=2aQvXBFPaq_zTlTfekczrUAAAAAk&t=ct%3Dns%26unitnum%3D2%26raptor%3Dcondor%26pos%3Dtop%26test%3D0
[5] https://www.theregister.com/2025/10/23/ai_model_bias_inevitable/
[6] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44aQvXBFPaq_zTlTfekczrUAAAAAk&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[7] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33aQvXBFPaq_zTlTfekczrUAAAAAk&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[8] https://youtu.be/_V4SxY5Kqfk?si=STOgNwnsb3GmkmcK&t=175
[9] https://www.scientificamerican.com/article/police-facial-recognition-technology-cant-tell-black-people-apart/
[10] https://www.bloomberg.com/graphics/2023-generative-ai-bias/
[11] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44aQvXBFPaq_zTlTfekczrUAAAAAk&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[12] https://montrealethics.ai/the-28-computer-vision-datasets-used-in-algorithmic-fairness-research/
[13] https://www.cnbc.com/2023/11/18/facebook-parent-meta-breaks-up-its-responsible-ai-team.html
[14] https://ai.meta.com/blog/responsible-ai-connect-2024/
[15] https://facet.metademolab.com/
[16] https://www.theregister.com/2025/11/05/oak_ridge_125m_funding/
[17] https://www.theregister.com/2025/11/05/attackers_experiment_with_gemini_ai/
[18] https://www.theregister.com/2025/11/05/amazon_perplexity_comet_legal_threat/
[19] https://www.theregister.com/2025/11/04/uk_court_getty_stability_ai/
[20] https://www.nature.com/articles/s41586-025-09716-2
[21] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33aQvXBFPaq_zTlTfekczrUAAAAAk&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[22] https://www.whitehouse.gov/wp-content/uploads/2025/07/Americas-AI-Action-Plan.pdf
[23] https://x.com/alicexiang/status/1986102965544108361
[24] https://whitepapers.theregister.com/
The [1]Fair Human-Centric Image Benchmark (FHIBE, or "Fee-bee") "is the first publicly available, consensually collected, and globally diverse fairness evaluation dataset for a wide variety of human-centric computer vision tasks," [2]according to Sony AI .
"A common misconception is that because computer vision is rooted in data and algorithms, it's a completely objective reflection of people," explains Alice Xiang, global head of AI Governance at Sony Group and lead research scientist for Sony AI, in [3]a video about the benchmark release. "But that's not the case. Computer vision can warp things depending on the biases reflected in its training data."
[4]
AI models, one way or another, [5]present bias , which perhaps with work can be minimized. Computer vision models may exhibit bias by inappropriately classifying people with specific physical characteristics in terms of occupation or some other label. They may categorize female doctors as nurses, for example.
[6]
[7]
According to Xiang, there have been instances in China where facial recognition systems on mobile phones have mistakenly allowed the device owner's family members to unlock the phone and make payments, an error [8]she speculates could come from a lack of images of Asian people in model training data or from undetected bias in the model.
"Much of the [9]controversy around facial recognition technologies has centered on their potential to be biased, leading to wrongful arrests, security breaches, and other harm," Xiang told The Register in an email. "GenAI models have also been [10]shown to be biased , reflecting harmful stereotypes."
[11]
There are a variety of [12]benchmark testing datasets for assessing the fairness of computer vision models. Meta, which in 2023 [13]disbanded its Responsible AI division but still likes to talk about [14]AI safety , has a computer vision benchmark called [15]FACET (FAirness in Computer Vision EvaluaTion).
But Xiang says the images that went into prior data sets were not gathered with the consent of those depicted – claims echoed in various AI copyright lawsuits now winding their way through the courts.
"The vast majority of computer vision benchmark datasets were collected without consent, and some were collected with consent but provide little information about the consent process, are not globally diverse, and are not suitable for a wide variety of computer vision tasks," she said.
[16]Oak Ridge lab bags $125M to bolt quantum onto supercomputers
[17]Attackers abuse Gemini AI to develop 'Thinking Robot' malware and data processing agent for spying purposes
[18]Amazon complains that Perplexity's agentic shopping bot is a terrible customer
[19]UK judge delivers a 'damp squib' in Getty AI training case, no clear precedent set
As detailed in [20]research published in Nature , the majority of the 27 evaluated computer vision datasets "were scraped from Internet platforms or derived from scraped datasets. Seven well-known datasets were revoked by their authors and are no longer publicly available."
The Sony AI authors also say that their image annotations (bounding boxes, segmentation masks, and camera settings, etc.) provide more detail, thus making FHIBE more useful to model developers.
[21]
FHIBE consists of "10,318 consensually-sourced images of 1,981 unique subjects, each with extensive and precise annotations." The images were gathered from more than 81 countries/regions.
The Sony AI researchers say they've used FHIBE to confirm that some vision models make less accurate predictions for those using "She/Her/Hers" pronouns due to the variability of hairstyles in related images. They also found that a model, when asked a neutral question about a person's occupation, would sometimes respond in a way that reinforced stereotypes by associating certain demographic groups with criminal activities.
The goal of the project, the AI group says, is to encourage more ethical and responsible practices for data gathering, use, and management.
Presently, in the US, there's not much federal government support for such sentiment. The Trump administration's "America's AI Action Plan"
[22]PDF
, released in July, makes no mention of ethics or fairness. Xiang, however, said, "The EU AI Act and some AI regulations in US states incentivize or require bias assessments in certain high-risk domains."According to Xiang, Sony business units have employed FHIBE in fairness assessments as part of their broader AI ethics review processes, in compliance with Sony Group AI Ethics Guidelines.
"FHIBE not only enables developers to audit their AI systems for bias but also shows that it is feasible to implement best practices in ethical data collection, particularly around consent and compensation for data rightsholders," said Xiang in [23]a social media post .
"At a time when data nihilism is increasingly common in AI, FHIBE strives to raise the standards for ethical data collection across the industry."
Asked to elaborate on this, Xiang said, "By 'data nihilism,' I mean the industry belief that data for AI development cannot be sourced consensually and with compensation, and that if we want cutting-edge AI technologies, we need to give up these data rights.
"FHIBE doesn't fully solve this problem since there's still the scalability issue (FHIBE is a small evaluation dataset, not a large training dataset), but one of our goals was to inspire the R&D community and industry to invest more care and funding into ethical data curation. This is an incredibly important problem – arguably one of the biggest problems in AI now – but far less attention is paid to innovation on the data layer compared to the algorithmic layer." ®
Get our [24]Tech Resources
[1] http://fairnessbenchmark.ai.sony
[2] https://ai.sony/articles/Groundbreaking-Fairness-Evaluation-Dataset-From-Sony%20AI%20/
[3] https://youtu.be/_V4SxY5Kqfk?si=NeuPdDwWYML4QLdJ&t=13
[4] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=2&c=2aQvXBFPaq_zTlTfekczrUAAAAAk&t=ct%3Dns%26unitnum%3D2%26raptor%3Dcondor%26pos%3Dtop%26test%3D0
[5] https://www.theregister.com/2025/10/23/ai_model_bias_inevitable/
[6] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44aQvXBFPaq_zTlTfekczrUAAAAAk&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[7] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33aQvXBFPaq_zTlTfekczrUAAAAAk&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[8] https://youtu.be/_V4SxY5Kqfk?si=STOgNwnsb3GmkmcK&t=175
[9] https://www.scientificamerican.com/article/police-facial-recognition-technology-cant-tell-black-people-apart/
[10] https://www.bloomberg.com/graphics/2023-generative-ai-bias/
[11] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44aQvXBFPaq_zTlTfekczrUAAAAAk&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[12] https://montrealethics.ai/the-28-computer-vision-datasets-used-in-algorithmic-fairness-research/
[13] https://www.cnbc.com/2023/11/18/facebook-parent-meta-breaks-up-its-responsible-ai-team.html
[14] https://ai.meta.com/blog/responsible-ai-connect-2024/
[15] https://facet.metademolab.com/
[16] https://www.theregister.com/2025/11/05/oak_ridge_125m_funding/
[17] https://www.theregister.com/2025/11/05/attackers_experiment_with_gemini_ai/
[18] https://www.theregister.com/2025/11/05/amazon_perplexity_comet_legal_threat/
[19] https://www.theregister.com/2025/11/04/uk_court_getty_stability_ai/
[20] https://www.nature.com/articles/s41586-025-09716-2
[21] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33aQvXBFPaq_zTlTfekczrUAAAAAk&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[22] https://www.whitehouse.gov/wp-content/uploads/2025/07/Americas-AI-Action-Plan.pdf
[23] https://x.com/alicexiang/status/1986102965544108361
[24] https://whitepapers.theregister.com/