In partnership with
Hi, I’m Danielle Turner, Welcome to the Property AI Tools newsletter where I share weekly deep dives into AI topics, the latest AI tools and news, all specifically for real estate.
If you’re at a roadblock and don’t know where to start with using AI in your real estate business, schedule a 1-1 discovery call.
Today I’ll be exploring:
The best and worst case scenarios that can arise when AI goes rogue
Tips to mitigate risk
The importance of being proactive to reduce the likelihood of high risk events
LATEST TECH NEWS
📰 Linkedin makes improvements to AI search
You can now search by describing your ideal career or asking specific questions in search.
📰 ChatGPT launches group chats
Give your team the power to have group conversations. Coordinate plans, conduct research and make decisions with up to 20 team members.
📰 Chinese manufacturer completes delivery of first humanoid robot
UBTech plans to deliver 500 robots by the year end.
Looking for unbiased, fact-based news? Join 1440 today.
Join over 4 million Americans who start their day with 1440 – your daily digest for unbiased, fact-centric news. From politics to sports, we cover it all by analyzing over 100 sources. Our concise, 5-minute read lands in your inbox each morning at no cost. Experience news without the noise; let 1440 help you make up your own mind. Sign up now and invite your friends and family to be part of the informed.
As AI becomes more embedded in critical business systems, the question of ‘Who’s accountable when things go wrong?’ arises. Whether it’s a minor incident or a major error with financial and legal implications, it’s important to understand the potential risks and the measures you can take to minimise the likelihood of mistakes getting out of hand.
Below i’ll be exploring three potential scenarios and the best vs worst outcomes. Take these scenarios and have a think about your own organization and the potential mistakes your AI systems could make in day to day operations.
When AI Spreads False or Defamatory Info About your Company
Generative AI systems like ChatGPT, Claude and Gemini generate and summarize information that can sometimes be incorrect, misleading or defamatory. This includes statements which are presented as facts, despite being untrue. If someone uses one of these tools to research your company and the information is incorrect, the situation could play out like this:
Best Case: The error is spotted quickly, the AI social listening tool you deployed to monitor brand mentions across platforms including ChatGPT has picked up the misleading statements and alerted your AI team. You were able to correct the information before it spread.
Worst Case: The misinformation was picked up by the press and cited in an article, causing lasting reputational damage. Legal action will be required to fix the mess.
Internal AI System Gives Incorrect Property Valuation
You’ve deployed a new Automated Valuation Model AI system to generate property valuations on autopilot. The system is new, still needs to be adjusted to suit your needs but is good enough to start using. You’ve decided to check every 10 valuations produced for quality control and have left the rest to be sent straight to clients.
Best Case: Your quality control surveyor picks up on a valuation that is wildly inaccurate, the property has non-standard characteristics that require an expert eye to value. They correct the error, update the valuation and send it to the client.
Worst Case: The incorrect valuation goes unnoticed and the client has uses this information to make financial decisions about their property. The client is unhappy and files a complaint causing reputational damage and a formal review.
Customer Facing AI Misadvises Client
Your customer service chatbot has given a customer incorrect advice, they’ve been told a property is available for showings when it’s actually already been sold.
Best Case: Your internal guardrail has flagged this error and a human has intervened in the conversation, quickly correcting the mistake.
Worse Case: The customer shows up to view the property, only to be told that it’s no longer on the market. The miscommunication from the chatbot has made your company look silly and they’ve since told all of their friends about the incident.
Precedent: Take a look at this recent case where Air Canada’s AI chatbot misled a a customer on it’s bereavement policy. The customer sued in court and won - Air Canada AI Vs Customer.
Mitigating the Risks
Conduct thorough risk assessments: Evaluate your AI systems before they are deployed for internal or external use. Pre-empt the potential risks, biases and flaws in the system as part of your AI risk management process so you can stay ahead of the curve.
Keep data up to date: Ensure your website is kept up to date and remove any invalid or outdated information. Remember, your website is the main source for AI to discover your brand and scrape relevant information.
Use RAG: Enable your AI system to retrieve information from approved, relevant, trusted sources to reduce the risk of misinformation. Find out more about this in my post here: What is Retrieval Augmented Generation (RAG)?
Implement internal guardrails with human oversight: Use tags to detect words, phrases and inputs in your AI system that are red flags. Have a person in post who is responsible for monitoring this system to mitigate risks and errors.
Establish a clear protocol: Have a clear procedure to follow when risk events occur and set chain of accountability to ensure responsible escalation.
For a deeper insight into the risks associated with AI, read the article Top AI Risks Every Real Estate Professional Should Know where this topic is covered in more detail.
Conclusion
AI presents huge opportunities for real estate businesses to innovate and drive growth, but it also presents risks which may impact your reputation, finances and client relationships; risks which cannot be ignored.
The examples above make one thing clear, being proactive is essential for responsible AI adoption. If you want to proactively mitigate that risk, you need to partner with experts who understand the regulatory landscape and technology to help you to navigate AI implementation with confidence.
If you’re ready to start taking these risks seriously, schedule a discovery call with me to discuss how I can help you form your strategic AI roadmap, assess your risk exposure and discover the best AI tools to ensure you’re prepared for the best and worst case scenarios.
Thanks for reading!
Would you like to sponsor this newsletter?
Email [email protected] to request our media kit