In Mexico City, Trust in AI and Trust in Data Protection Move Together

US · CanadaAmérica Latina

Weekly AI Watch

Every week, what changed and what it means for you. No hurry, no alarm.

Week of 23 October 2026EN | ESPast editions18,276 papers found · 17,064 read · 5,349 belonged · 37 stories passed every check

Observatorio de la
Inteligencia Artificial

Cada semana, qué cambió y qué significa para usted. Sin prisa y sin alarma.

Semana del 23 de octubre de 2026EN | ESEdiciones anteriores18,276 estudios encontrados · 17,064 leídos · 5,349 correspondían · 37 notas pasaron todas las comprobaciones

weeklyAI · Week of 23 October 2026weeklyAI · Semana del 23 de octubre de 2026

← Your rights← Sus derechos

survey · Frontiers in artificial intelligence · la publicación, 13 may 2026 · gratis

Quien confía en la IA también cree que sus datos están mejor cuidados

Un estudio preguntó a 101 profesionistas de la Ciudad de México por su opinión sobre la inteligencia artificial y sobre la protección de sus datos personales. Midió opiniones, no lo que las empresas o las autoridades hacen en realidad.

Versión breve · la versión detallada sigue, unos 6 min

Pregunte al chat de weeklyAI

Pregúnteme por este estudio: a quiénes se estudió, qué encontró y qué no dice.

Las conversaciones se guardan mientras exista weeklyAI, para mejorar la publicación. Se responde en el idioma en que usted escribe.

El estudio, de un vistazo
Quiénes: Profesionistas con estudios universitarios
Cuántos: 101 participantes
Dónde: Ciudad de México, México
Cuándo: 2025
Tipo de estudio: survey
Quién lo hizo: Universidad San Ignacio de Loyola, la UNAM y la Universidad Tecnológica del Perú
El límite que importa: Solo son opiniones recogidas una vez, por internet, entre personas con carrera universitaria.

Por eso el 0.635 habla de confianza, y la confianza se parece más a una promesa que a un recibo.
Lectura de weeklyAI

Ver · 0:25

Piense en dos preguntas que se contestan en la misma encuesta, una tras otra: qué tanto confía usted en la inteligencia artificial, y qué tan protegidos cree que están sus datos personales. En la Ciudad de México, las respuestas a esas dos preguntas se movieron juntas.

Los investigadores reunieron a 101 personas con estudios universitarios que vivían en la capital en 2025 y les aplicaron un cuestionario de 24 preguntas. La relación entre las dos opiniones fue de 0.635. Es una cifra que va de 0 a 1: mientras más alta, más de la mano van las dos respuestas. Aquí, quienes calificaban mejor a la IA también calificaban mejor la protección de sus datos.

La relación apareció en los cuatro temas sobre protección de datos que el cuestionario separó. El vínculo más fuerte fue con la idea de que existen mecanismos que resguardan la información (0.669), y siguió el de la confianza en que las instituciones cumplen (0.648). El más débil fue el conocimiento de las leyes (0.380). Son las cifras de una de las dos medidas que usaron los investigadores; la otra dio valores parecidos y en el mismo orden.

Ese último dato dice algo. Algunos encuestados confiaban en la inteligencia artificial sin conocer del todo sus derechos sobre los datos. La ley mexicana de 2010 para el sector privado fijó esos derechos: acceso, rectificación, cancelación y oposición.

Conviene ver lo que este número no es. Fueron opiniones recogidas una sola vez, por internet, entre personas con carrera universitaria. El estudio no revisó si las empresas cumplen la ley ni si alguien sufrió una filtración de datos. Tampoco dice nada sobre otros países, ni sobre quienes no tienen estudios universitarios.

México actualizó en 2025 su ley de protección de datos para el sector privado, con reglas sobre el procesamiento automático. Y su autoridad de datos llegó a ordenar a Google México borrar los datos de salud de un ciudadano. Son decisiones reales de ese tipo las que el cuestionario no midió.

Por eso el 0.635 habla de confianza, y la confianza se parece más a una promesa que a un recibo. Lo que puede pedir donde vive es concreto: busque si la autoridad de datos de su país publica sus resoluciones, si las empresas avisan de las filtraciones y si los reclamos tienen respuesta. Ahí se ve si la confianza se gana o solo se declara.

Qué significa para usted

Ese 0.635 mide opiniones, no lo que las empresas o las autoridades hacen con sus datos. Si usted vive en México, puede revisar si la autoridad de datos publica sus resoluciones y si las empresas avisan cuando sufren una filtración. En otros países, vale la pena buscar qué dice su propia ley sobre el uso de su rostro, su voz y su información.

Medrano-Sánchez, E. J., Ruiz-Ramírez, E., Ayllon, M. L. (2026). Between innovation and risk: artificial intelligence and data protection in digital Mexico. Frontiers in Artificial Intelligence. https://doi.org/10.3389/frai.2026.1716108

Quién pagó: Los autores declararon que no recibieron apoyo financiero para este trabajo ni para su publicación, y declararon no tener relaciones comerciales o financieras que pudieran constituir un conflicto de interés.

Los hallazgos de otros estudios que aquí se mencionan los conocemos por este documento, que fue el que leímos; no abrimos cada uno de esos estudios.

Versión detalladaLos pasajes copiados del artículo, las ilustraciones y cada fuente con cuánto leímos de ella · 1240 palabras · unos 6 minLeerla →Cerrar

Confiar en la IA y creer que sus datos están protegidos van juntos, según un sondeo en la Ciudad de México

Un cuestionario a 101 profesionistas con estudios universitarios encontró que quienes ven mejor la inteligencia artificial también ven mejor la protección de sus datos personales. El estudio midió opiniones, no lo que las empresas y autoridades hacen en realidad.

En la Ciudad de México, un grupo de 101 profesionistas con estudios universitarios contestó en 2025 un cuestionario de 24 preguntas sobre dos cosas: qué opinan de la inteligencia artificial y qué opinan de la protección de sus datos personales. Cada bloque tenía 12 preguntas, y cada bloque se dividía en cuatro temas de tres preguntas. Los autores encontraron que quienes calificaban mejor la inteligencia artificial también calificaban mejor la protección de sus datos¹. El vínculo apareció en los cuatro temas que midieron: conocimiento de las leyes, riesgos percibidos, cumplimiento y responsabilidad de las instituciones, y percepción de resguardos.

Las respuestas se ordenaron en tres niveles: malo, regular y bueno. De los 35 participantes que calificaron mal la inteligencia artificial, 25 calificaron mal la protección de sus datos, y ninguno la calificó bien. De los 23 que calificaron bien la inteligencia artificial, 15 calificaron bien la protección. La relación más débil fue con el conocimiento de las leyes: entre quienes calificaron mal la IA, solo dos dijeron conocer bien las normas. La más fuerte fue con la percepción de resguardos técnicos y administrativos.

Los propios autores anotan una consecuencia de esa diferencia: parte de los encuestados confiaba en la IA sin estar del todo informada sobre las leyes de privacidad². El estudio no puede decir si esa confianza nace de la información o de la costumbre. Tampoco puede decir que la IA proteja mejor los datos de nadie: es una sola foto tomada en un momento, con 101 personas reclutadas por internet y de manera voluntaria, y los mismos autores advierten que quienes más usan la tecnología pudieron ser más propensos a responder³. No se recogieron edad, género ni sector profesional. Y todo lo medido son opiniones, no prácticas reales de empresas ni de autoridades. Los autores publicaron en Frontiers in Artificial Intelligence, sin financiamiento, y declararon no tener conflictos de interés. Trabajan en la Universidad San Ignacio de Loyola, la UNAM y la Universidad Tecnológica del Perú.

Piense en el aviso que aparece cuando instala una aplicación: una pantalla con letra chica y un botón de "aceptar". Ahí se juega buena parte de lo que este estudio pregunta. En México, la protección de datos personales de empresas privadas se regula formalmente desde 2010, con la Ley Federal de Protección de Datos Personales en Posesión de los Particulares, que fijó los derechos de acceso, rectificación, cancelación y oposición⁴. En 2025 se promulgó una actualización sustancial que refuerza obligaciones, crea mecanismos de vigilancia y aborda de manera explícita el tratamiento automatizado y el análisis masivo de datos⁵.

La ley mexicana es reciente y todavía se está consolidando; frente al reglamento europeo quedan huecos, sobre todo en la aplicación uniforme de principios como la minimización y la portabilidad de los datos⁶. El caso de Google México sirve para ver cómo funciona esto en la práctica: el Instituto Federal de Acceso a la Información y Protección de Datos obligó a la empresa a borrar los datos de salud de un ciudadano, con criterios tomados de una sentencia del tribunal europeo en el caso Google Spain contra la agencia española de protección de datos⁷. Es decir, hubo una resolución concreta y un dato borrado.

Lo curioso es que la misma tecnología que crea el problema también sirve para vigilarlo. Los sistemas de inteligencia artificial y aprendizaje automático detectan anomalías y previenen filtraciones analizando grandes volúmenes de datos⁸. Entre los peligros más citados están las filtraciones de datos sensibles, la inyección de código malicioso y la elaboración intrusiva de perfiles de usuarios⁹.

El tamaño del fenómeno ayuda a dimensionarlo: 35 por ciento de las organizaciones ya emplea inteligencia artificial, otro 42 por ciento evalúa usarla, y al menos 250 millones de empresas en el mundo ya la usan o la investigan¹⁰. Imagine esa cifra como una empresa por cada habitante de un país entero varias veces mayor que México, cada una decidiendo qué hace con los datos que recibe.

Fuera de México, el panorama regulatorio se mueve. El reglamento europeo de protección de datos exige transparencia y pone límites al tratamiento automatizado, y la ley de privacidad de California establece derechos parecidos en Estados Unidos¹¹. La próxima ley europea de inteligencia artificial busca imponer obligaciones específicas a quienes desarrollan y usan sistemas inteligentes¹². En América Latina, varios países adoptaron normas inspiradas en el modelo europeo, pero la falta de armonización entre estados, junto con iniciativas como el reconocimiento de los neuroderechos en Chile, dibuja un mosaico regulatorio cuya eficacia depende en buena medida de la coordinación entre países¹³. Lo que usted puede exigir en su país, y a quién, depende de esa pieza.

Así lo leemos nosotros. La confianza en una herramienta y la confianza en quien la vigila no son lo mismo, y el estudio las encontró pegadas. Cuando alguien dice que la IA le parece buena, puede estar diciendo dos cosas: que entiende cómo funciona, o que se acomodó a que no hay otra opción y le llama protección a lo disponible. El estudio no distingue entre las dos. Nos haría pensar que estamos equivocados si quienes más confían en la IA fueran también quienes más conocen sus derechos y más han reclamado ante una empresa o una autoridad. Eso no se midió aquí.

Hay una segunda lectura posible. La confianza entre personas se sostiene cuando lo que alguien promete coincide con lo que hace. Si una empresa o una autoridad anuncia que cuida sus datos y luego se documenta un incumplimiento claro, esa confianza debería bajar. Si no baja, es señal de que estaba puesta en el discurso y no en los resultados. Usted puede ponerla a prueba: guarde el aviso de privacidad de un servicio que use, anote qué promete sobre sus datos y compare esa promesa con lo que ocurra. Y busque si la autoridad de datos de su país ha emitido resoluciones o sanciones reales, y si alguna se parece a un caso suyo. Que la institución exista no es lo mismo que haya atendido casos como el suyo.

Los autores recomiendan, en línea con el Objetivo de Desarrollo Sostenible 16 de la ONU, fortalecer los marcos institucionales con regulación, transparencia y educación digital, impulsar la colaboración entre sector público y privado, y fomentar la vigilancia y la alfabetización digital para lograr una confianza informada¹⁴. El estudio no probó que esas medidas funcionen; solo las propone.

La pregunta que queda abierta es la que el propio estudio no pudo responder: si la confianza que la gente pone en la IA se sostendría al conocer mejor sus derechos. Los autores señalan que la evidencia empírica en América Latina sigue siendo escasa, y que las capacidades técnicas y la producción académica se concentran en pocos países.

Ese es el punto práctico para usted. La próxima vez que una aplicación le pida sus datos, pregúntese si está de acuerdo con lo que hace o solo se acomodó a que no hay otra opción. Anote qué le gustaría poder exigir y a quién. Con eso en la mano, la pantalla de "aceptar" deja de ser un trámite y se vuelve una decisión suya.

De dónde sale cada dato de contexto, y cuánto leímos de cada documento

Medrano-Sánchez, E. J., Ruiz-Ramírez, E., Ayllon, M. L. (2026). Between innovation and risk: artificial intelligence and data protection in digital Mexico. Frontiers in Artificial Intelligence. https://doi.org/10.3389/frai.2026.1716108 - el artículo del que trata esta nota — el artículo completo — el pasaje: "Results indicated that a more favorable perception of AI was associated with more favorable perceptions of personal data protection and its dimensions. The correlations demonstrated a moderate and significant positive association ( p < 0.001) between perceptions of AI and overall perceptions of data protection."
Medrano-Sánchez, E. J., Ruiz-Ramírez, E., Ayllon, M. L. (2026). Between innovation and risk: artificial intelligence and data protection in digital Mexico. Frontiers in Artificial Intelligence. https://doi.org/10.3389/frai.2026.1716108 - el artículo del que trata esta nota — el artículo completo — el pasaje: "the correlation coefficient between the assessment of AI and regulatory knowledge was lower than that observed with the general perception of protection, indicating that part of the sample trusted AI without necessarily being fully informed about privacy laws"
Medrano-Sánchez, E. J., Ruiz-Ramírez, E., Ayllon, M. L. (2026). Between innovation and risk: artificial intelligence and data protection in digital Mexico. Frontiers in Artificial Intelligence. https://doi.org/10.3389/frai.2026.1716108 - el artículo del que trata esta nota — el artículo completo — el pasaje: "the study relied on a convenience sample composed exclusively of professionals with university-level education. Given the digital mode of recruitment and the voluntary nature of participation, it cannot be ruled out that individuals with higher familiarity or engagement with digital technologies and artificial intelligence may have been more likely to respond."
Medrano-Sánchez, E. J., Ruiz-Ramírez, E., Ayllon, M. L. (2026). Between innovation and risk: artificial intelligence and data protection in digital Mexico. Frontiers in Artificial Intelligence. https://doi.org/10.3389/frai.2026.1716108 - el artículo del que trata esta nota — el artículo completo — el pasaje: "In the Mexican context, personal data protection for private-sector entities has been formally regulated since the enactment of the Federal Law on the Protection of Personal Data Held by Private Parties (LFPDPPP) in 2010, which established core principles, obligations, and the ARCO rights framework (access, rectification, cancellation, and opposition) for data subjects"
Medrano-Sánchez, E. J., Ruiz-Ramírez, E., Ayllon, M. L. (2026). Between innovation and risk: artificial intelligence and data protection in digital Mexico. Frontiers in Artificial Intelligence. https://doi.org/10.3389/frai.2026.1716108 - el artículo del que trata esta nota — el artículo completo — el pasaje: "a substantial regulatory update was enacted in 2025, introducing strengthened obligations, oversight mechanisms, and provisions explicitly addressing automated processing and advanced data analytics"
Medrano-Sánchez, E. J., Ruiz-Ramírez, E., Ayllon, M. L. (2026). Between innovation and risk: artificial intelligence and data protection in digital Mexico. Frontiers in Artificial Intelligence. https://doi.org/10.3389/frai.2026.1716108 - el artículo del que trata esta nota — el artículo completo — el pasaje: "In Mexico, privacy legislation is relatively recent and still undergoing consolidation; significant gaps remain compared to the European GDPR, especially regarding uniform application of principles such as minimization and portability"
Medrano-Sánchez, E. J., Ruiz-Ramírez, E., Ayllon, M. L. (2026). Between innovation and risk: artificial intelligence and data protection in digital Mexico. Frontiers in Artificial Intelligence. https://doi.org/10.3389/frai.2026.1716108 - el artículo del que trata esta nota — el artículo completo — el pasaje: "Mexico's Federal Institute for Access to Information and Data Protection (IFAI) compelled Google Mexico to erase a citizen's health data, relying on criteria established by the Court of Justice of the European Union in the Google Spain vs. AEPD case"
Medrano-Sánchez, E. J., Ruiz-Ramírez, E., Ayllon, M. L. (2026). Between innovation and risk: artificial intelligence and data protection in digital Mexico. Frontiers in Artificial Intelligence. https://doi.org/10.3389/frai.2026.1716108 - el artículo del que trata esta nota — el artículo completo — el pasaje: "The same AI and machine learning technologies that pose new challenges also provide unprecedented capabilities for anomaly detection and breach prevention by analyzing large datasets"
Medrano-Sánchez, E. J., Ruiz-Ramírez, E., Ayllon, M. L. (2026). Between innovation and risk: artificial intelligence and data protection in digital Mexico. Frontiers in Artificial Intelligence. https://doi.org/10.3389/frai.2026.1716108 - el artículo del que trata esta nota — el artículo completo — el pasaje: "Among the most cited dangers are leaks of sensitive data ( He et al., 2025 ), malicious code injection ( Pelekis et al., 2025 ), and intrusive user profiling"
Medrano-Sánchez, E. J., Ruiz-Ramírez, E., Ayllon, M. L. (2026). Between innovation and risk: artificial intelligence and data protection in digital Mexico. Frontiers in Artificial Intelligence. https://doi.org/10.3389/frai.2026.1716108 - el artículo del que trata esta nota — el artículo completo — el pasaje: "Currently, 35% of organizations employ AI, another 42% are evaluating its use, and by 2024 more than half plan to integrate it into their processes; in absolute terms, at least 250 million companies are already using or investigating AI technologies"
Medrano-Sánchez, E. J., Ruiz-Ramírez, E., Ayllon, M. L. (2026). Between innovation and risk: artificial intelligence and data protection in digital Mexico. Frontiers in Artificial Intelligence. https://doi.org/10.3389/frai.2026.1716108 - el artículo del que trata esta nota — el artículo completo — el pasaje: "Europe's General Data Protection Regulation (GDPR) mandates transparency and limits on automated data processing ( Al-Tkhayneh et al., 2023 ), while the California Consumer Privacy Act (CCPA) establishes similar rights within the United States ( Burton et al., 2024 )."
Medrano-Sánchez, E. J., Ruiz-Ramírez, E., Ayllon, M. L. (2026). Between innovation and risk: artificial intelligence and data protection in digital Mexico. Frontiers in Artificial Intelligence. https://doi.org/10.3389/frai.2026.1716108 - el artículo del que trata esta nota — el artículo completo — el pasaje: "the forthcoming European Union AI Act aims to impose specific obligations on developers and users of intelligent systems"
Medrano-Sánchez, E. J., Ruiz-Ramírez, E., Ayllon, M. L. (2026). Between innovation and risk: artificial intelligence and data protection in digital Mexico. Frontiers in Artificial Intelligence. https://doi.org/10.3389/frai.2026.1716108 - el artículo del que trata esta nota — el artículo completo — el pasaje: "Several Latin American countries have adopted regulations inspired by the GDPR, but the lack of harmonization among states, coupled with unique initiatives like the recognition of neuro-rights in Chile, reveals a regulatory mosaic whose effectiveness depends significantly on interstate coordination"
Medrano-Sánchez, E. J., Ruiz-Ramírez, E., Ayllon, M. L. (2026). Between innovation and risk: artificial intelligence and data protection in digital Mexico. Frontiers in Artificial Intelligence. https://doi.org/10.3389/frai.2026.1716108 - el artículo del que trata esta nota — el artículo completo — el pasaje: "recommendations aligned with SDG 16 include strengthening institutional frameworks (regulation, transparency, digital education), fostering public-private collaboration, and promoting digital oversight and literacy to achieve informed trust"

Los hallazgos de otros estudios que aquí se mencionan los conocemos por este documento, que fue el que leímos; no abrimos cada uno de esos estudios.

survey · Frontiers in artificial intelligence · the paper, 13 May 2026 · free

In Mexico City, Trust in AI and Trust in Data Protection Move Together

Researchers asked 101 university-educated professionals in Mexico City how they see AI and how they see their data protection. The two opinions moved together; whether anyone's data was safer, the study did not measure.

Short version · the longer version follows, about 6 min

Ask the weeklyAI chat

Ask me about this study: who was studied, what it found, and what it does not say.

Conversations are saved for as long as weeklyAI exists, to improve the publication. Answers come in the language you write in.

The study at a glance
Who: university-educated professionals
How many: 101
Where: Mexico City
When: 2025
Kind of study: survey
Who did it: Universidad San Ignacio de Loyola, Universidad Nacional Autónoma de México, Universidad Tecnológica del Perú
The limit that matters: Opinions only, one snapshot, recruited online by convenience; not proof anyone's data was safer.

A high trust score is not evidence that your information is safe.
weeklyAI's reading

In 2025, researchers asked 101 university-educated residents of Mexico City two sets of questions. One set asked how they see artificial intelligence: whether they understand it, use it, trust it, find it ethical. The other asked how they see the protection of their personal data: whether they know the law, feel at risk, believe institutions comply, trust the safeguards.

The answers lined up. People who rated AI more favorably also rated data protection more favorably. The overall link came out at 0.635, which the researchers call moderate to high. A link of 0 would mean the two opinions have nothing to do with each other, and a link of 1 would mean they move in perfect step.

The link held across all four parts of data protection. The strongest was trust in safeguards, at 0.669 by one measure and 0.617 by the other. The weakest was knowledge of the law itself, at 0.380 by one measure and 0.338 by the other.

That weakest number is worth a second look. The article says part of the sample trusted AI without necessarily being fully informed about privacy laws. Mexico's 2010 law set out four rights over personal data: access, correction, cancellation and objection.

Keep in mind what was measured. These were opinions, collected once, from a convenience sample recruited online in one city. The study did not check whether any company or authority actually protected anyone's data. A high trust score is not evidence that your information is safe.

The scale of the questions matters too. It says nothing about people outside Mexico City, or about people with less schooling or less familiarity with digital tools. The authors also collected no age or gender, so the picture has little detail behind it.

Mexico has been moving on the rules themselves. It updated its private-sector data protection law in 2025, adding provisions on automated processing, and its regulator once ordered Google Mexico to erase a citizen's health data. Those are concrete decisions. The study measured none of them.

So the 0.635 describes how two attitudes travel together, not a guarantee that anyone is better protected. Watch whether the institutions that hold your data publish audits, breach notices and the outcomes of complaints. That is where trust would have to be earned.

What this means for you

The study describes a link among 101 people in one city who answered questions once, and it cannot say whether any company or authority actually protected anyone's data. Watch whether the institutions that hold your information publish audits, breach notices and the outcomes of complaints, and look at what the law where you live says about automated processing.

Who paid: The authors declared that no financial support was received for this work or its publication, and they declared no commercial or financial relationships that could be a conflict of interest.

The longer versionThe passages copied from the paper, the pictures, and every source with how much of it we read · 1275 words · about 6 minRead it →Close

In Mexico City, People Who Trust AI More Also Think Their Data Is Safer

A survey of 101 university-educated professionals found the two opinions move together. It cannot show that either one is true.

Consider the moment a company asks to scan your face before it lets you in, or an app wants your voice to unlock a service. You agree, or you don't. What decides it is rarely the written policy. It is whether you believe the people behind the system will handle what they take.

A survey published in May in *Frontiers in Artificial Intelligence* asked 101 university-educated professionals living in Mexico City how they felt about artificial intelligence and how they felt about the protection of their personal data. The two sets of answers lined up: those who viewed AI more favorably also tended to view their data as better protected¹.

The authors describe this as people carrying their trust in the technology over to the laws and institutions meant to guard their information².

The study used a 24-item questionnaire, half of it about AI and half about data protection, answered on a five-point scale from strongly disagree to strongly agree. The AI half asked about understanding the technology, experience using it, trust in it, and whether it behaves ethically. The data-protection half asked about knowledge of the law, perceived risk, confidence that institutions comply, and confidence in safeguards. All 101 respondents answered at one point in 2025.

The link with knowledge of the law was weaker than the link with the overall sense of data protection. That gap matters: the authors write that part of the sample trusted AI without necessarily being well informed about privacy law³. In other words, some respondents felt protected without being able to say by what.

The study cannot tell us whether trust in AI causes anyone to be better protected, or the reverse. It is one snapshot, taken once, with no follow-up. It also cannot tell us whether the respondents' data was in fact handled well. It measured opinions, not what companies or authorities did.

The 101 people were recruited online, by convenience, and all had a university education. The authors themselves warn that people more familiar with digital technology may have been likelier to answer⁴. That makes the group a poor stand-in for Mexico as a whole, or for any other country.

Now put the finding beside the law those respondents live under. Mexico has regulated the handling of personal data held by private companies since 2010, through a federal law that set out principles and a set of rights known by the initials ARCO: access, rectification, cancellation and opposition⁵. In 2025 the rules were substantially updated, with stronger obligations, new oversight, and provisions that speak directly to automated processing and advanced data analysis⁶. So the people surveyed were answering while their legal framework was itself in motion.

The framework still has gaps. The study's authors describe Mexican privacy law as recent and still consolidating, with significant distance from Europe's General Data Protection Regulation, particularly in applying principles such as collecting only the data you need and moving it between services⁷. Elsewhere in the region the picture is uneven: several Latin American countries have adopted GDPR-inspired rules, but they are not harmonized with one another, and Chile has gone its own way in recognizing so-called neuro-rights, according to the authors' review⁸.

A Mexican case shows what the rights can do when someone uses them. The country's Federal Institute for Access to Information and Data Protection, known as IFAI, ordered Google Mexico to erase a citizen's health data, drawing on criteria set by the Court of Justice of the European Union in the Google Spain case⁹. That is a right exercised and a company made to act: the sword, not just the shield.

The comparison with other jurisdictions is where the interests sit. Europe's GDPR requires transparency and sets limits on automated processing, and California's privacy law establishes similar rights for residents of that state¹⁰. A European Union AI Act is still forthcoming and aims to place specific obligations on those who build and those who use intelligent systems¹¹. Mexico's rules, by the authors' account, are younger and less settled than these.

The scale of what is being regulated helps explain the concern. By the figures the authors cite, 35 percent of organizations already use AI, another 42 percent are evaluating it, and at least 250 million companies worldwide are using or investigating it¹². The dangers most often named are leaks of sensitive data, malicious code injection, and intrusive profiling of users¹³. The same technologies, the authors note, can also help detect anomalies and prevent breaches by analyzing large datasets¹⁴: the tool cuts both ways.

Here is how we read it. The pattern is familiar from other walks of life: people tend to defer to whoever sounds like the expert or the official, and they keep going along because stopping feels like defying a legitimate authority. If that is what is happening here, the warm feelings about AI may be an echo of confident institutional voices rather than a judgment about safeguards anyone has actually checked. We would expect that to hold in homes across the region, not only in Mexico City: a person who finds a new system useful may assume the rules around it are working, without ever testing them. You could tell we are wrong if people who rate AI highly can also name the specific protections they have personally looked into, or if their confidence survives a well-publicized failure to protect data.

There is a second way to see it, and it worries us more. The side that holds the data and the technical tools ends up knowing far more about ordinary people than they know about it, and that imbalance quietly becomes an imbalance of power. A warm feeling about AI can sit alongside very little real leverage. A respondent may believe protection exists while lacking the practical means to see what is held about them or to challenge it, so their confidence could outrun the rights they actually hold. You would know this reading is too dark if people who feel good about AI can readily demand access to their records or their deletion and get a timely, clear answer.

What can you do with either reading? Find out which specific request you can make to a company that holds your data, and make it once. Ask what they hold, or ask them to delete it. Note how quickly and how clearly they respond. That single exchange will tell you more about your protection than any survey of attitudes, including this one.

The authors' own recommendations run toward institutions rather than individuals: stronger regulation, transparency, digital education, collaboration between the public and private sectors, and oversight and literacy aimed at what they call informed trust¹⁵. Notice what that phrase concedes. Trust is the goal, but it is meant to be earned by things a person can check.

We also read work on transparency in AI law, and it offers a caution worth carrying. In the European rules studied there, transparency appears as an intuitive but elusive idea, with no firm legal definition. That is not a verdict on this survey. It is a reminder that "transparent" on a company's page may mean very little until someone defines it and someone enforces it.

The face-scanning prompt, the app that wants your voice: you will meet one again soon. When you do, the useful question is not whether the system seems trustworthy. It is who would answer if it went wrong, and what you could check yourself before you say yes.

Where each piece of context comes from, and how much of it we read

Medrano-Sánchez, E. J., Ruiz-Ramírez, E., Ayllon, M. L. (2026). Between innovation and risk: artificial intelligence and data protection in digital Mexico. Frontiers in Artificial Intelligence. https://doi.org/10.3389/frai.2026.1716108 - the article this story is about — the whole article — the passage: "Results indicated that a more favorable perception of AI was associated with more favorable perceptions of personal data protection and its dimensions. The correlations demonstrated a moderate and significant positive association ( p < 0.001) between perceptions of AI and overall perceptions of data protection."
Medrano-Sánchez, E. J., Ruiz-Ramírez, E., Ayllon, M. L. (2026). Between innovation and risk: artificial intelligence and data protection in digital Mexico. Frontiers in Artificial Intelligence. https://doi.org/10.3389/frai.2026.1716108 - the article this story is about — the whole article — the passage: "the surveyed population tended to transfer the trust placed in AI systems to the legal and technical structures that protect their personal information, so that technological perception functioned as a relevant factor associated with citizens' judgment regarding security and responsibility in data processing"
Medrano-Sánchez, E. J., Ruiz-Ramírez, E., Ayllon, M. L. (2026). Between innovation and risk: artificial intelligence and data protection in digital Mexico. Frontiers in Artificial Intelligence. https://doi.org/10.3389/frai.2026.1716108 - the article this story is about — the whole article — the passage: "the correlation coefficient between the assessment of AI and regulatory knowledge was lower than that observed with the general perception of protection, indicating that part of the sample trusted AI without necessarily being fully informed about privacy laws"
Medrano-Sánchez, E. J., Ruiz-Ramírez, E., Ayllon, M. L. (2026). Between innovation and risk: artificial intelligence and data protection in digital Mexico. Frontiers in Artificial Intelligence. https://doi.org/10.3389/frai.2026.1716108 - the article this story is about — the whole article — the passage: "the study relied on a convenience sample composed exclusively of professionals with university-level education. Given the digital mode of recruitment and the voluntary nature of participation, it cannot be ruled out that individuals with higher familiarity or engagement with digital technologies and artificial intelligence may have been more likely to respond."
Medrano-Sánchez, E. J., Ruiz-Ramírez, E., Ayllon, M. L. (2026). Between innovation and risk: artificial intelligence and data protection in digital Mexico. Frontiers in Artificial Intelligence. https://doi.org/10.3389/frai.2026.1716108 - the article this story is about — the whole article — the passage: "In the Mexican context, personal data protection for private-sector entities has been formally regulated since the enactment of the Federal Law on the Protection of Personal Data Held by Private Parties (LFPDPPP) in 2010, which established core principles, obligations, and the ARCO rights framework (access, rectification, cancellation, and opposition) for data subjects"
Medrano-Sánchez, E. J., Ruiz-Ramírez, E., Ayllon, M. L. (2026). Between innovation and risk: artificial intelligence and data protection in digital Mexico. Frontiers in Artificial Intelligence. https://doi.org/10.3389/frai.2026.1716108 - the article this story is about — the whole article — the passage: "a substantial regulatory update was enacted in 2025, introducing strengthened obligations, oversight mechanisms, and provisions explicitly addressing automated processing and advanced data analytics"
Medrano-Sánchez, E. J., Ruiz-Ramírez, E., Ayllon, M. L. (2026). Between innovation and risk: artificial intelligence and data protection in digital Mexico. Frontiers in Artificial Intelligence. https://doi.org/10.3389/frai.2026.1716108 - the article this story is about — the whole article — the passage: "In Mexico, privacy legislation is relatively recent and still undergoing consolidation; significant gaps remain compared to the European GDPR, especially regarding uniform application of principles such as minimization and portability"
Medrano-Sánchez, E. J., Ruiz-Ramírez, E., Ayllon, M. L. (2026). Between innovation and risk: artificial intelligence and data protection in digital Mexico. Frontiers in Artificial Intelligence. https://doi.org/10.3389/frai.2026.1716108 - the article this story is about — the whole article — the passage: "Several Latin American countries have adopted regulations inspired by the GDPR, but the lack of harmonization among states, coupled with unique initiatives like the recognition of neuro-rights in Chile, reveals a regulatory mosaic whose effectiveness depends significantly on interstate coordination"
Medrano-Sánchez, E. J., Ruiz-Ramírez, E., Ayllon, M. L. (2026). Between innovation and risk: artificial intelligence and data protection in digital Mexico. Frontiers in Artificial Intelligence. https://doi.org/10.3389/frai.2026.1716108 - the article this story is about — the whole article — the passage: "Mexico's Federal Institute for Access to Information and Data Protection (IFAI) compelled Google Mexico to erase a citizen's health data, relying on criteria established by the Court of Justice of the European Union in the Google Spain vs. AEPD case"
Medrano-Sánchez, E. J., Ruiz-Ramírez, E., Ayllon, M. L. (2026). Between innovation and risk: artificial intelligence and data protection in digital Mexico. Frontiers in Artificial Intelligence. https://doi.org/10.3389/frai.2026.1716108 - the article this story is about — the whole article — the passage: "Europe's General Data Protection Regulation (GDPR) mandates transparency and limits on automated data processing ( Al-Tkhayneh et al., 2023 ), while the California Consumer Privacy Act (CCPA) establishes similar rights within the United States ( Burton et al., 2024 )."
Medrano-Sánchez, E. J., Ruiz-Ramírez, E., Ayllon, M. L. (2026). Between innovation and risk: artificial intelligence and data protection in digital Mexico. Frontiers in Artificial Intelligence. https://doi.org/10.3389/frai.2026.1716108 - the article this story is about — the whole article — the passage: "the forthcoming European Union AI Act aims to impose specific obligations on developers and users of intelligent systems"
Medrano-Sánchez, E. J., Ruiz-Ramírez, E., Ayllon, M. L. (2026). Between innovation and risk: artificial intelligence and data protection in digital Mexico. Frontiers in Artificial Intelligence. https://doi.org/10.3389/frai.2026.1716108 - the article this story is about — the whole article — the passage: "Currently, 35% of organizations employ AI, another 42% are evaluating its use, and by 2024 more than half plan to integrate it into their processes; in absolute terms, at least 250 million companies are already using or investigating AI technologies"
Medrano-Sánchez, E. J., Ruiz-Ramírez, E., Ayllon, M. L. (2026). Between innovation and risk: artificial intelligence and data protection in digital Mexico. Frontiers in Artificial Intelligence. https://doi.org/10.3389/frai.2026.1716108 - the article this story is about — the whole article — the passage: "Among the most cited dangers are leaks of sensitive data ( He et al., 2025 ), malicious code injection ( Pelekis et al., 2025 ), and intrusive user profiling"
Medrano-Sánchez, E. J., Ruiz-Ramírez, E., Ayllon, M. L. (2026). Between innovation and risk: artificial intelligence and data protection in digital Mexico. Frontiers in Artificial Intelligence. https://doi.org/10.3389/frai.2026.1716108 - the article this story is about — the whole article — the passage: "The same AI and machine learning technologies that pose new challenges also provide unprecedented capabilities for anomaly detection and breach prevention by analyzing large datasets"
Medrano-Sánchez, E. J., Ruiz-Ramírez, E., Ayllon, M. L. (2026). Between innovation and risk: artificial intelligence and data protection in digital Mexico. Frontiers in Artificial Intelligence. https://doi.org/10.3389/frai.2026.1716108 - the article this story is about — the whole article — the passage: "recommendations aligned with SDG 16 include strengthening institutional frameworks (regulation, transparency, digital education), fostering public-private collaboration, and promoting digital oversight and literacy to achieve informed trust"

The findings of other studies mentioned here are known to us through this document, which is the one we read; we did not open each of those studies.

The free weekly drip by email

One email a week: the free story of every section, with its picture, and the headlines of the rest.

No payment, nothing to cancel. Unsubscribe from any issue.

El goteo semanal gratis por correo

Un correo por semana: la nota gratuita de cada sección, con su imagen, y los titulares de las demás.

Sin pago y sin nada que cancelar. Puede darse de baja desde cualquier envío.