Quick wins for a faster PC:
Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Clear out junk files and repair common Windows errorsFree Scan →In October 2023, Mercado Libre began a pilot using VidMob’s Diversity and Inclusion Scoring tool to inspect some advertising creatives before publication. The system was described as assessing visually inferred age range, gender and skin tone against criteria set by Mercado Libre. It was a representation audit—not proof that an ad was culturally inclusive, that delivery was equitable, or that the campaign improved business or social outcomes.
What Mercado Libre announced
Exame reported on October 16, 2023, that Mercado Libre was testing artificial intelligence on part of its advertising campaigns. The stated objective was to turn diversity, equity and inclusion commitments into a repeatable marketing process and to help identify stereotypes and “hegemonic” viewpoints before ads went live. (Exame)
This was presented as a pilot, not a permanent company-wide system with published performance results. VidMob described Mercado Libre as the first global company to adopt the product; that is a statement from the supplier, not an independently verified industry ranking. (Ecossistema Inova)
How the pilot was supposed to work
- Set the criteria: Mercado Libre defined the diversity and inclusion parameters it wanted to check.
- Submit the creative: Selected video and social-media assets were sent for analysis before publication.
- Analyze visible people: VidMob’s system examined representational attributes in the imagery.
- Produce a report or score: The output gave the creative team a repeatable signal about the representation it contained.
- Revise when necessary: Teams could change the creative before it was released.
- Escalate complex cases: VidMob said human curation was available when a higher level of interpretive detail was needed. (Ecossistema Inova)
The design is best understood as pre-flight quality control. It could make a manual review faster and more consistent, but it did not make the system an autonomous judge of whether an advertisement was “diverse.”
Outdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchPC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11#1 Best Overall
What attributes did the AI examine?
Public descriptions of the first phase named three categories:
- Apparent age range;
- Gender;
- Skin tone.
“Apparent” is important. These are attributes the system attempted to infer from what appears in an image or video, not verified information about a person’s identity, self-description or demographic record. The available accounts do not establish reliable measurement of race, ethnicity, disability, sexuality, body type, religion or socioeconomic status.
Representation is not the same as inclusion
The pilot addressed who appears in creative, not every way an advertising system can be fair or inclusive. Four different questions are often collapsed into one:
| Question | What it concerns | What the 2023 pilot establishes |
|---|---|---|
| Creative representation | Who is visible and how often | Yes, within the reported age, gender and skin-tone checks |
| Narrative inclusion | Agency, roles, language, stereotypes and cultural context | Not established by a visual score |
| Audience delivery | Who receives, sees or can access the ad | Not measured by this creative audit |
| Business or social response | Clicks, sales, brand lift, attitudes or reduced stereotyping | No public evidence from the pilot |
An ad can show a numerically varied cast while assigning everyone stereotyped roles, using tokenism or excluding people through its language and design. Conversely, a narrowly focused cast can be appropriate for a specific product or story. A score is therefore an audit signal, not a certification of inclusion.
Why use AI for a diversity review?
For a regional advertiser producing many versions of video and social creative, automated screening could offer several operational benefits:
- Reviewing more assets quickly than a purely manual process;
- Applying brand-defined checks consistently across campaigns;
- Creating a recordable baseline for visible representation;
- Flagging omissions while there is still time to revise an ad;
- Giving creative, media and DEI teams a common review step.
Those are plausible reasons to deploy a scoring system. They are not measured outcomes of Mercado Libre’s pilot. The public accounts do not report that the tool improved conversion, brand lift, representation or campaign efficiency.
Why human review remained necessary
Many consequential DEI judgments are contextual rather than visual. A person reviewing a flagged creative may need to ask:
- Who has agency in the story?
- Are people shown in varied and consequential roles, or only as background props?
- Does the wording reinforce a stereotype?
- Does the portrayal make sense in the local cultural context?
- Is the treatment respectful in every market where the ad will run?
VidMob’s description of human curation acknowledges that classification alone cannot answer those questions. A responsible workflow should also let reviewers see confidence levels, understand why an asset was flagged and record when a human accepts or overrides a recommendation.
Accessibility was proposed as a later capability
VidMob said a future version would examine accessibility-related elements, including choices such as color and font. Exame reported VidMob’s claim that accessibility improvements could potentially increase campaign reach by as much as 20 percent. That figure was a supplier estimate, not a measured uplift from Mercado Libre’s pilot, and it should not be reported as proof that Mercado Libre gained 20 percent more reach. (Exame)
Accessibility also extends beyond contrast and typography. Depending on the format, a meaningful review may include captions, audio description, sign-language interpretation, flashing content, reading order, clear voiceover and platform-specific controls. The cited reporting does not show which of these were implemented.
The main technical and ethical risks
Visual inference is not demographic truth
Age, gender and skin tone can be difficult to infer from lighting, makeup, image quality, camera angle or partial visibility. Gender in particular should not be treated as a simple binary property discovered by a camera.
Uneven model performance
A computer-vision model can behave differently across skin tones, lighting conditions, occlusion, image quality and cultural settings. The public material provides no accuracy, calibration or subgroup error rates for this deployment.
Tokenism and false reassurance
Optimizing a count can encourage teams to add visible diversity without changing roles, narrative power or product positioning. A favorable number may also create the impression that DEI review is complete when language and context remain unchecked.
Regional variation
Standards that work in Brazil may not capture local understandings of identity, colorism, indigenous representation, language or disability in Mexico, Argentina or another market. VidMob said the project would start in Brazil and potentially extend to 18 other Latin American countries; the cited sources do not verify that the full rollout occurred. (Ecossistema Inova)
Privacy and consent questions
The available reports do not explain retention periods, processing locations, model-training use, access controls or consent practices for imagery containing faces. Those are material questions for any advertiser putting creative assets through an external computer-vision service.
Creative review is not delivery fairness
Checking who appears in an ad does not reveal whether an advertising platform distributes that ad evenly across demographic groups. Algorithmic delivery fairness is a separate technical and policy problem, examined in research such as this study of demographic disparities in ad delivery.
The Tool Desk
Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →What the public record still does not show
| Unanswered question | Status |
|---|---|
| How many ads were analyzed? | Not published in the cited accounts |
| What were baseline and final scores? | Not published |
| How many campaigns were changed? | Not published |
| What were accuracy, false-positive and false-negative rates? | Not published |
| How often did humans override the system? | Not published |
| Did the pilot change conversion, sales or brand lift? | No evidence reported |
| Did accessibility improve? | No outcome data reported |
| Was rollout completed across 19 markets? | Not independently verified |
A separate Mercado Libre AI project in 2026
In April 2026, Mutt Data described Brand ID, a multimodal system for broader brand-compliance review of image and video ads, using capabilities such as speech-to-text and language models. It is a different use case from the 2023 Diversity and Inclusion Scoring pilot. The later case study does not prove that the diversity pilot succeeded or that Mercado Libre still uses the same scorer. (Mutt Data)
What enterprise buyers should ask before adopting a similar tool
- Are categories clearly defined, and are they treated as visual estimates rather than identity facts?
- Can the buyer set different criteria for each local market?
- Does the system explain scores and provide confidence thresholds?
- How does it handle crowds, small or obscured faces, animation, illustrations, synthetic people and repeated appearances?
- Does it analyze language, audio and narrative context, or only visible faces?
- Is human review built into the workflow, with an audit trail for overrides?
- What data is retained, where is it processed and is it used to train models?
- Does the product address creative representation, audience delivery, campaign outcomes—or only the first of these?
VidMob is the directly relevant supplier in the 2023 announcement, but current package names, pricing and 2026 availability for that specific diversity product were not established. The practical value of such a system depends less on a headline score than on validation, local expertise and the quality of the human decisions that follow it.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
