Artificial Intelligence Guideline

• Guidelines

  BCFSA’s Guidelines provide a practical application of the information and give suggested best practice guidance to assist real estate professionals. These guidelines provide BCFSA’s interpretation of the Real Estate Services Act (“RESA”) and all other applicable legislation.

  In addition, BCFSA’s Guidelines may be a useful information source for the general public looking for information about standards of conduct for real estate professionals.

Purpose

The purpose of this Guideline is to help you understand the potential risks associated with using artificial intelligence (“AI”) while providing real estate services. Using AI does not exempt you from your legal and ethical obligations. Licensees remain fully accountable for their actions and must uphold the standards established by the regulatory framework, and all other applicable legislation and requirements. This Guideline is designed to alert licensees to specific areas of concern that demand careful consideration when deciding whether to implement AI applications in your real estate practices.

1. Exercise caution when using generative AI systems
2. Protect client information when using AI
3. Ensure the accuracy of information produced by AI
4. Understand systemic bias in AI systems

Guidelines

Applicable Sections of Legislation

Section 28 Real Estate Services Rules, Managing broker responsibilities

Section 30 Real Estate Services Rules, Duties to clients

Section 33 Real Estate Services Rules, Duty to act honestly

Section 34 Real Estate Services Rules, Duty to act with reasonable care and skill

Section 40 Real Estate Services Rules, Advertising — Restrictions and requirements

Section 41 Real Estate Services Rules, False or misleading advertising prohibited

* Additional Rules and/or enactments may apply, depending on the facts.

Definitions

Artificial intelligence (“AI”) is the information technology that performs tasks that would ordinarily require biological brainpower to accomplish, such as making sense of spoken language, learning behaviors, or solving problems.¹

Client means, in relation to a real estate licensee, the principal who has engaged the real estate licensee to provide real estate services to or on behalf of the principal.

Generative AI is a type of AI that can generate text, audio, code, videos, and images based on large volumes of data they were trained on, incremented with user inputs. Chatbots such as ChatGPT, Google Bard, and Microsoft Copilot are examples of generative AI tools.

Real estate advertising means any form of identification, promotion, solicitation, or representation relating to:

• Real estate;
• A trade in real estate;
• The provision of real estate services; and
• Including a sign or other notice relating to real estate, a trade in real estate, or the provision of real estate services.

Real estate services mean

• Rental property management services;
• Strata management services; or
• Trading services.

Publish, in relation to real estate advertising, includes:

• Causing or permitting real estate advertising to be published; and
• Displaying real estate advertising, or causing or permitting real estate advertising to be displayed.

1. Exercise Caution When Using Generative AI Systems

When providing real estate services, you will often be asked to draft contracts, modify standard forms, or provide advice to clients. When considering whether to use AI tools to help you with professional tasks, it is important to recognize that AI lacks professional licensure, and the platforms do not offer dependable warranties or representations. AI tools are designed to “guess” answers to questions they do not have complete information on and may end up telling you something that is not correct. Do not rely on AI-generated information without conducting a thorough review of the content. You could potentially be held accountable for any inaccuracies propagated by the AI.

AI output may also infringe upon the intellectual property rights of others; ownership of content generated by AI is unsettled at this time, so it is important to be cautious with AI’s outputs. Additionally, if you attempt to employ AI to provide advice in areas where your expertise is limited or where you lack proper licensure, you may be subject to regulatory sanctions.

Remember, the Real Estate Services Rules mandate that licensees advise clients to seek independent professional advice for matters beyond the licensees’ purview. Autonomous AI advice is not a substitute for the expertise of qualified professionals.

Advertising Considerations

Another consideration relates to how you may use AI tools to organize, refine, and post advertising photos. As AI technology improves, it is likely that more sophisticated tools will become available to enhance the marketability of various pictures and videos (e.g., removing imperfections, filling rooms with virtual designer décor and/or fixtures, modernizing the space, etc.). However, you must be careful not to make false representations in your advertising. While it may be okay to use an AI tool to virtually stage a house, you should label photos or videos that have been altered and enhanced so you do not mislead a potential buyer. Licensees may not create any advertising that is false or misleading – caution and clear labeling is imperative when utilizing AI technology for any kind of marketing.

Although it is acceptable to direct an AI tool to assist with a task or to provide information, you should always carefully review the documentation or information generated by the AI before relying on the information or sharing it with clients.

2. Protect Client Information When Using AI

When providing real estate services, you will often handle sensitive client information, including personal information. Maintaining the confidentiality of this information is a fundamental duty you owe to your clients under the Real Estate Services Act (“RESA”) and applicable privacy law. For this reason, it is important that you exercise great caution when using AI tools.

AI introduces many confidentiality and privacy considerations because of its capacity to process confidential and personal data from a large body of information, and its ability to data mine the queries users feed into them that might include sensitive information. For example, when engaging with an AI system, you should never assume that the data you input is safeguarded and private. This assumption could lead you to input confidential or personal information into the AI tool to get insights into market trends or analyze client preferences and behaviours. However, it is crucial to recognize that the processing of confidential or personal data by generative AI algorithms could inadvertently expose such information – even where the licensee has taken steps to anonymize their client’s information – by matching data from different sources to identify an individual. This arises from the fact that some AI platforms may store and analyze user interaction data to enhance the performance of their systems and refine their training models. As a result, AI systems might include sensitive data in their results by utilizing previous user inputs.

AI also introduces ethical considerations pertaining to ownership and data use. For instance, service providers are beginning to offer a broader range of AI-related services, such as reviewing strata documents and producing tailored reports. However, the service providers developing these tools could potentially be using that data for other purposes. For example, the incidental data from a strata document report could be of significant interest to various industries, including insurance companies and competing service providers, for purposes ranging from targeted marketing campaigns to risk assessment and competitive analysis. The prospect of AI service providers leveraging this data to sell insights to marketing firms or think tanks introduces complex ethical and privacy considerations.

Acquire Consent Before Using Client Information

Confidential client information includes things such as a person’s name, address, phone number, income, spending habits, etc. Always ensure that you have acquired your clients’ informed consent before using their information (and information obtained on their behalf) in an AI tool. You should also review the privacy disclosures of the AI service provider to ensure they do not put you or your client at risk of disclosing confidential information. Strata management licensees should ensure that the use of strata corporation information in an AI tool is reflected in their service agreement. The strata corporation should be advised to seek legal advice in this situation.

You should also refrain from using AI tools without your client’s express consent. Once you collect any personal information, you can only use it for the purposes that a reasonable person would consider appropriate in the circumstances. They must be consistent with the purposes you identified to the individual from whom you collected the personal information or with the consent of that individual unless an exception applies (see Section 14 of the Personal Information Protection Act for additional information). Review BCFSA’s Privacy Guidelines for more information.

Given these potential implications, it is paramount to exercise prudence and adhere to established data protection protocols, as well as brokerage policies and procedures when using generative AI. Unless you are certain that an AI tool does not store or utilize user data, you should avoid entering any confidential or personal information into an AI tool.

3. Ensure the Accuracy of Information Produced By AI

The accuracy of AI-generated results depends on the quality and relevance of the data it is trained on. If the foundational data is outdated, incomplete, or inaccurate, AI models may produce responses that are out-of-date or incorrect. Licensees who choose to use AI-generated content in their real estate practices remain accountable for the accuracy of any real estate advertising they publish or other representations they make in the course of providing real estate services.

Since AI systems may produce inaccurate results, it is important for you to verify any results from an AI tool before using them. For example, if you are using AI to write listing content like property descriptions, you must ensure the information is correct. Licensees must not publish any real estate advertising that they know, or ought to know, contains a false statement or misrepresentation. This applies not only to written text but to pictures as well.

Ultimately, before publishing any AI generated content, you should review and validate the results to detect and correct inaccuracies. You should also communicate to clients and the public when AI tools are being utilized and explain their potential impacts and limitations to maintain transparency. For example, if you use an AI chatbot or interface when you are unavailable, you should inform your client of the situation.

4. Understand Systemic Bias in AI Systems

Artificial intelligence systems often grapple with issues related to bias, which can carry significant ethical and societal consequences. One primary source of AI bias emerges from the data used for training. Historical data, reflecting societal prejudices, can introduce and perpetuate biases in AI systems. For example, AI systems employed for tenant screening may inadvertently rank individuals differently based on factors like race, gender, or language, thus reinforcing discriminatory practices.

To address bias concerns, it is important to thoroughly verify any AI-generated data. Ensure that any professional decisions you make based on AI-generated information are fair and unbiased, especially in sensitive areas like tenant screening. Rental property managers may also wish to review the Office of the Information and Privacy Commissioner’s guidance document to help them understand what the rules in B.C. are about what personal information landlords may collect from anyone seeking to enter into a tenancy agreement. Transparency regarding the use of AI is also crucial. You should communicate to clients, unrepresented parties, and any party whose information is harnessed (e.g., tenant applicants) when AI tools are utilized, explaining both the potential impact and limitations.

To mitigate potential negative consequences stemming from AI bias or discrimination, establish procedures for error/bias correction when AI-generated recommendations are involved. Ultimately, you bear the responsibility for any decisions you make using output from AI systems. A prudent licensee should exercise due diligence to ensure that AI outputs and decisions align with applicable laws and regulations.

Managing Broker Considerations

Managing brokers play a critical role in overseeing and guiding the activities of their licensees. When it comes to integrating AI systems into a brokerage’s operations or helping licensees successfully incorporate AI into their practice, there are several important considerations.

Brokerage Policies
Establish clear brokerage policies and procedures regarding the use of AI tools and how client information is handled. You may wish to implement protocols for data retention, encryption, and secure sharing of information within the brokerage. Emphasize the need to anonymize personally identifiable information before entering anything into AI systems.

Risk Mitigation
Incorporate risk mitigation strategies into your brokerage’s AI usage policy. Where your brokerage’s related licensees use AI-generated content, it should be used as a supplement to their own expertise rather than a replacement. Ensure that licensees know that they must refer clients to seek independent professional advice in areas outside of their expertise. Brokerages may also want to discuss the use of AI tools with their insurance provider to determine if it impacts their coverage.

Educate Your Licensees
Ensure that all your licensees are well informed about potential risks associated with using AI systems and the importance of protecting client information. Consider providing training on data privacy regulations, confidentiality, and the proper use of AI tools.

Vendor Due Diligence
If your brokerage is considering using third-party AI tools, thoroughly vet the vendors to ensure their data handling practices align with your brokerage’s privacy policies, client agreements, and obligations under the Real Estate Services Act (“RESA”). Review their terms of service and privacy agreements and discuss your AI usage with your clients.

Ongoing Supervision
Continuously monitor how AI tools are being used by your licensees. Regularly review AI-generated content to ensure it aligns with brokerage policies and maintains accuracy.

Managing broker supervision and oversight play a pivotal role in helping licensees stay on side of the RESA regulatory framework and other relevant legislation when using AI technology in the provision of real estate services.

Print