AI Just Saved a $1.4 Million Sale. The Next Agent Advantage Is Human.

By Tim, Julie, Dan, Keri, Chris, Kacie, Tammy and Orlando · September 11, 2026

🎧 We’re also on Apple Podcasts and Spotify!

A Pasadena agent built two financial models from a car in France this week and saved a $1.4 million sale for sellers who were ready to walk away. She isn't a programmer. She asked Claude for help.

That is the whole episode.

This week's PowerHouse Talk covers the moment AI stopped being a novelty and started becoming a real operating advantage: one photo turning into a walkable 3D home, $30,000 robo-taxis becoming a possible investment class, AI deciding which agents to recommend, and the uncomfortable truth that a beautiful listing presentation no longer proves anything.

The agents who win next will not be the ones who choose between technology and human connection. They'll be the ones who use technology to create more time for the conversations only humans can have.

The episode welcomes new host Tammy Irby, a Northern Virginia agent with 41 years in real estate and 13 children raised while building a top-producing career. Her best year ever came during one of the supposedly worst markets in recent memory, which makes her the perfect person to join this conversation.

Tammy's message to agents is simple: you're never too old to adapt, and the business is still one of the best careers available for anyone who wants flexibility, family time, and the ability to build something meaningful. She also arrived fresh from a three-day AI conference in Las Vegas, admitting that she came home both energized and humbled by how much she still had to learn.

That combination — decades of experience plus a willingness to learn new tools — is the exact profile the episode argues will win the next market.

Dan shared the first major case study. He had an FSBO seller with renters, three dogs, a lease running another six months, outdated cell-phone photos, and a complicated 1031-exchange timeline. His initial expectation was that the seller would need months of follow-up before becoming realistic enough to hire an agent.

Keri took the situation and fed it into Claire, her AI assistant. Within minutes, Claire produced:

The result: the seller took the listing offline 20 minutes after receiving the email and called immediately. The listing was signed before noon.

Keri's point was crucial: AI did not win the listing by itself. It gave her the time and analytical depth to show up with a complete plan. She was able to spend the day with her family while Claire did hours of research and strategy work in the background.

The panel spent time on a surprising observation: younger people are not automatically the most enthusiastic AI adopters. Tammy said her 21-year-old daughter, who is graduating college this year, is fluent in AI and treats it as normal. Her 35-year-old daughter is cautious because she worries AI will stop people from thinking for themselves. Her 45-year-old daughter is fully committed to learning it.

Dan compared the divide to college football's transfer portal and NIL system. Coaches who refused to adapt did not preserve the old system — they simply fell behind competitors who embraced the new one. His warning for real estate agents: agents who resist AI will eventually be unable to match the productivity, efficiency, and value delivered by agents who use it.

The panel's broader message was not that AI replaces agents tomorrow. It is that the gap between adopters and non-adopters is becoming exponential. One step taken with AI may now equal 50 steps taken manually. By the time a resistant agent has walked 50 feet, the AI-enabled agent may have taken 2,500 equivalent steps.

One of the most important discussions covered AI search and agent recommendations. Large language models are now recommending one to three agents per market, sometimes ranking a brand-new licensee above a veteran agent. The problem: AI may not have access to MLS production data, so it can make recommendations based on whatever information it can find online.

Dan's advice is practical: if AI does not recommend you, ask it what source data it used. It may be pulling from Reddit, blog posts, landing pages, Zillow directories, or other third-party pages.

The panel recently encountered a case where the top listing agent in a market failed to appear in AI search while an agent who had sold only two homes in 20 years did appear. The reason was Zillow's agent directory and its Premier Agent designation. When the AI's data sources changed, the results changed too.

The one consistent recommendation: make sure your Google Business Profile is complete and your actual production statistics are publicly searchable. If you want AI to identify you as a top listing agent, the data cannot remain hidden behind a brokerage login or a portal paywall. Put the facts online:

Keri shared the strategy her team is using to compete for higher-priced listings: creating long-form landing pages and case studies for every property they have sold over $2 million. Each page is optimized with relevant keywords, schema markup, internal linking, and detailed explanations of the problem, strategy, and result.

Her team is also creating case studies about homes that other agents failed to sell first. One example involved a property that had five previous listing attempts before Ker's team sold it for 101% of the asking price in five days.

The point is not to publish generic "market update" posts. It is to create searchable evidence that answers the exact questions sellers ask:

Keri said her website traffic increased 238% month over month after implementing a sustained AEO strategy. Her advice: use AI to produce the content efficiently, but make sure the content reflects real experience and real transactions.

Julie brought up a less obvious AI problem: writing can look perfect for several paragraphs and then suddenly shift tone, introduce an odd phrase, or even insert a random word in another language. She cited research suggesting that AI editing can create its own internal patterns as it repeatedly regenerates and improves text, causing the middle of a piece to drift away from the original voice.

The practical rule is simple: every piece of AI-generated writing still needs a final human edit.

That applies to:

Tim explained that agents can build a "humanizer" tool into their AI workflow, but Julie's point remains: if the content has been edited or regenerated multiple times, run the entire piece through a final review rather than assuming the initial humanizer pass is enough.

The episode took a turn into robo-taxis and their possible impact on real estate investing. The discussion centered on Tesla's proposed cyber-cab model, reportedly priced around $30,000, with businesses potentially able to buy fleets and place them into an autonomous ride network.

The panel discussed a possible model where a car owner earns a share of the fare revenue while Tesla or the operating network handles dispatching, charging, cleaning, and fleet management. One calculation discussed on the show suggested that 200 vehicles could potentially generate $1 million in annual net income, although the panel presented that as a speculative scenario rather than a guaranteed return.

The real estate implications are bigger than transportation:

The panel's optimism was not really about cars. It was about the changing definition of an asset. A $30,000 autonomous vehicle may become more like a revenue-producing rental unit than a depreciating personal expense — if the business model works as described.

Tammy raised one of the most practical issues in the episode: sellers constantly ask why they aren't receiving feedback after showings. Her answer: no feedback is feedback.

The absence of feedback may mean:

Tammy's point was that agents should not become dependent on other agents' opinions. Feedback is often biased, especially in multiple-offer situations. The more reliable source is the market itself: showings, saves, inquiries, offers, days on market, and the rate at which comparable homes are selling or failing to sell.

That is why the panel repeatedly returned to the 180-day seller communication plan available through their coaching system. The purpose is to give sellers a consistent stream of useful updates instead of waiting for random feedback from other agents.

The panel's strongest philosophical point was that AI should not replace the human experience. It should create more time for it.

Tammy gave an example of using AI to rewrite a listing description for five personality types: amiable, analytical, driver, influencer, and communicator. The result is not just prettier copy. It is messaging designed to appeal to more buyers and drive more traffic into the home.

Keri shared an even more ambitious idea: using AI-generated cinematic video to show not just an empty dining room, but the memories that could be created there. Her team is also using AI avatars to tell the stories behind previous sales, including cases where other agents tried and failed before her team closed the deal.

When asked whether people care if content is AI-generated, the panel's answer was consistent: audiences want to be entertained, educated, or empowered. They care about the result more than the production method. Platforms may automatically label AI content, but that does not necessarily reduce its value if the content is useful and credible.

Dan described the shift from social media to interest media. People increasingly see content based on what they watch and engage with, not simply from the people they follow. That means posting constantly is no longer enough. Agents need content that earns attention, and they may need to pay to put that content in front of specific audiences.

Tammy delivered the clearest conclusion: as AI makes everyone's descriptions, flyers, videos, and websites look similar, the ability to speak confidently with clients becomes more valuable, not less.

Kacie's $16.5 million listing from the previous episode was cited again as proof. The listing came from a hairstylist referral, not a polished online brand. The client still interviewed multiple agents, and Kacie won by explaining her marketing and compensation confidently rather than discounting herself.

The panel's rule for the AI era:

AI will not eliminate the grind. It will put rocket boosters on it.

The agents who use AI well will have more time, better preparation, faster content production, deeper client insights, and more opportunities to serve people. The agents who refuse to adapt will continue doing the same work manually while competitors move faster and produce more.

But the work that remains most valuable will be unmistakably human:

The final instruction from the episode was simple: optimize for optimism. Surround yourself with people, tools, and ideas that pull you forward. Use AI to become the disruptor of your own business before someone else disrupts it for you.

— Tim, Julie, Dan, Keri Chris, Kacie, Tammy and Orlando
Hosts, Power House Talk

whylibertas.com/harris
whylibertas.com/dan
whylibertas.com/heller
whylibertas.com/kacie-anderson
whylibertas.com/orlando-montiel

FOR REAL ESTATE PROFESSIONALS READY TO LEVEL UP

You already know you can do more. Now you'll know exactly what to do.

Start Premier Coaching for $1 for your first seven days.

***

💬 Want your question featured on an upcoming episode? Reply to this email or DM us on Instagram— we might break it down next.

📩 Know an agent who should be in this room?
Forward them Power House Talk.

Pioneer, the summit where CX leaders redefine what’s possible, is on October 7th.

Join leaders from Fin, Anthropic, Clay, and Kalshi for an insightful conversation on the state of AI transformation.

You’ll discover how some of the most innovative minds in CX have transformed their organizations, learn how they think about CX, and hear how they're planning for what's next.

Join the conversation in San Francisco, or tune in virtually.

Reserve your spot