Code of Ethics for the Use of Artificial Intelligence in Marketing and Sales
Introduction
Artificial intelligence is transforming how companies conduct market research, generate content, segment audiences, qualify prospects, personalize offers, and serve their customers. These capabilities can increase productivity and improve the customer experience, but they can also facilitate deceptive, invasive, discriminatory, or manipulative practices when used without clear principles.
An AI governance plan defines who may use AI, for what purposes, and under which controls. A code of ethics serves a complementary function: it establishes the values that should guide decisions, even when an internal policy or regulation does not yet address a specific situation.
This code provides a practical framework for companies to use artificial intelligence in marketing and sales in a responsible, transparent, and customer-centered manner. Its guiding principle is that no improvement in conversion, efficiency, or revenue justifies deceiving consumers, exploiting their vulnerabilities, invading their privacy, or transferring responsibility for a business decision to an algorithm.
Developed under the direction of Dr. Carlos Valdez, based on a comparative analysis of proposals produced by ChatGPT, Claude, and Gemini, with support from ChatGPT in the integration and editorial drafting process.
Principle 1 — Customer Dignity Above Conversion
The company must recognize the customer as a person with autonomy and decision-making capacity, not merely as a data profile or revenue opportunity.
AI should be used to understand and serve the customer more effectively, not to exploit fears, anxiety, financial urgency, inexperience, or emotional vulnerability. A campaign should not be considered successful solely because it increased conversion. It should also be evaluated according to whether it respected the customer’s freedom to understand the offer and make a decision without undue pressure.
This means avoiding false urgency, fabricated scarcity, excessive automated persistence, deceptive interfaces, and messages deliberately designed to weaken consumer judgment. Personalization should help customers find relevant solutions, not become a form of individualized manipulation.
Principle 2 — Truthfulness and the Prohibition of Commercial Hallucinations
All information generated or communicated through AI must be accurate, verifiable, and supported by evidence.
AI systems must not invent product features, prices, discounts, availability, certifications, expected results, testimonials, or contractual terms. When an AI system does not have a reliable answer, it should acknowledge its limitation and transfer the inquiry to a person or an authorized source.
The company assumes responsibility for any commercial hallucination produced by the systems it selects and uses. The statement “the AI generated it” does not eliminate the responsibility to verify information before communicating it.
Campaigns, proposals, quotations, and communications related to benefits, prices, or terms must undergo human review before they are published or sent.
Principle 3 — Transparency and Authenticity
Customers have the right to know when they are interacting with artificial intelligence and when they are communicating with a person.
Every chatbot, sales assistant, or voice agent must clearly identify itself at the beginning of the interaction. The company must also provide a visible option for requesting human assistance when the inquiry or decision requires it.
Transparency also applies to synthetic content. AI-generated images, voices, videos, and avatars must not be used to make audiences believe that a person made a statement, used a product, or endorsed a brand when that did not occur.
The company must maintain internal traceability for content generated or substantially modified through AI. Labels, watermarks, metadata, or digital content credentials should be used when required by regulation or when there is a reasonable risk of confusion.
Principle 4 — Privacy and Responsible Use of Data
Prospect and customer data must be treated as information entrusted to the company for a specific purpose, not as a resource available for any future use.
The company should process only the information necessary to provide a service, recommendation, or personalized experience. It should not develop intrusive psychological profiles or infer sensitive characteristics unless those practices are legitimate, necessary, and proportionate.
Data obtained during a purchase, conversation, or digital interaction should not automatically be used to train models or support new campaigns without a legitimate basis and without considering customer expectations.
Entering customer databases, contracts, business strategies, or confidential information into public generative AI tools that have not been authorized by the company’s technology, privacy, and security functions must also be prohibited.
Principle 5 — Fairness and Algorithmic Nondiscrimination
Systems that segment audiences, qualify prospects, recommend products, or determine prices must not be considered neutral simply because they operate automatically.
The company must evaluate whether these systems create unjustified differences among customers or groups in comparable situations. Apparently neutral variables, such as ZIP code, device type, language, or digital behavior, may act as proxies for protected characteristics and produce indirect exclusion.
AI must not be used to deny opportunities, reduce service quality, or impose disadvantageous conditions for reasons unrelated to legitimate commercial criteria.
Lead-scoring systems should support professional judgment, not replace it entirely. Salespeople and managers must be able to review, question, and correct their recommendations.
Principle 6 — Ethical Pricing and Personalization
Artificial intelligence may be used to adjust prices, promotions, and offers when variations are based on legitimate factors such as demand, inventory, seasonality, costs, volume, or service level.
It must not be used to increase prices or reduce benefits by exploiting personal vulnerabilities, individual urgency, or demographic characteristics.
Customers in comparable circumstances should not receive materially different treatment without a legitimate business justification. Systems must also operate within previously authorized limits, and any significant exception must be reviewed by a responsible person.
An algorithmic recommendation may increase profit margins while still being ethically inappropriate. Commercial efficiency does not justify practices that erode customer trust.
Principle 7 — Human Oversight and Accountability
The greater the financial, emotional, reputational, or contractual impact of an interaction, the greater the level of human oversight required.
A serious complaint, high-value negotiation, contractual dispute, or decision that significantly affects a customer should not be left entirely in the hands of an automated system.
AI may assist through analysis, summaries, and recommendations, but the final decision must remain with a person who has the authority and ability to consider the broader context.
Every system must have an identifiable person responsible for it. The company cannot attribute responsibility for a decision to an algorithm or dilute accountability among departments and providers. Agencies and third parties that use AI on behalf of the organization must comply with the same ethical standards.
Principle 8 — Security, Correctability, and Continuous Monitoring
Every AI system that interacts with customers or processes commercial information must be protected against manipulation, unauthorized access, fraud, and use beyond its established limits.
The company must be able to immediately correct, restrict, or deactivate any system that behaves improperly. Higher-risk systems must include an emergency shutdown mechanism and an alternative manual process for continuing to serve customers.
Oversight does not end when the system is launched. A system that operates correctly today may lose accuracy, reproduce new biases, or change its behavior as data, the model, or the market evolves.
For this reason, the company must periodically review response quality, customer complaints, differences in treatment among customer segments, and any unexpected outcomes. The ethical performance of an AI system is not certified once; it must be sustained through continuous monitoring.
Application of the Code
This code should be incorporated into provider selection, tool approval, campaign design, employee training, and performance evaluation.
Every user of AI in marketing and sales must understand the system’s purpose, the data it uses, the actions it can perform, its limitations, and the procedure for stopping it.
Managers must evaluate AI initiatives across four dimensions simultaneously: commercial value, quality of the customer experience, level of risk, and effect on customer trust.
When there is uncertainty regarding the legitimacy of a particular use, the organization should temporarily suspend it and submit it for review. Speed of execution must never take precedence over accountability.
Expected Outcome
A company that adopts this code is not merely seeking regulatory compliance. It is seeking to develop a commercial culture in which artificial intelligence is used with sound judgment, discipline, and respect for people.
The expected outcome is an organization capable of innovating without losing control of its decisions, protecting customers without undermining productivity, and using automation to strengthen sustainable commercial relationships.
Ethics is not an obstacle to effective marketing and sales. It is the condition that makes such effectiveness legitimate, trustworthy, and sustainable over time.
Author: Dr. Carlos Valdez Date: Summer 2026
Editorial assistance: ChatGPT 5.5 — analysis, drafting, proofreading, grammatical editing, research, and verification of academic references.
Image generated with: ChatGPT 5.5
Publication: Revista Mercadotecnia y Ventas
© 2026 Revista Mercadotecnia y Ventas. All rights reserved.
Reproduction without the author’s permission is prohibited.
Editorial syndication: This content is available for syndication. For editorial licensing or collaboration opportunities, contact: carlos.valdez@mercadotecniayventas.com
