CRE's Adoption of AI - article + my question to you all

Here is a super indepth and insightful article from CRE Analyst on survey insights about our industry's adoption of and outlook on AI. Worth a read, or at least a skim of the takeaways.

https://creanalyst.substack.com/p/ai-in-cre-survey-mythbusting-the?utm_id=849251116&utm_medium=paid-social&utm_source=linkedin

One topic the article covers is the extent CRE professionals have woven AI into their work. Apparently 76% use it as a basic assistant, but only 5% use it to automate their workflow. My firm uses enterprise ChatGPT and I've been trying to integrate it into my work (asset management at a MF). It increases my productivity in maybe 50-60% of areas, sometimes in big ways (e.g. summarization, drafting emails, market research, Excel templates from scratch) and often in only marginal ways (e.g. regular reporting, legal doc analysis, CAM Rec review).

All in all it's been hard to really integrate it into my workflow but I feel like it could free me up by 70%+ if I could just have it understand my ask and pointed at the data I need. I just go in circles with ChatGPT until it starts hallucinating or refusing my reiterated requests. My goal over the next 3-9 months is to become a true power user: automate my monthly property updates (which comes from various sources - a hodgepodge of email threads and OneNote notes), tenant and/or property financials review, and probably most importantly, organizing my to-do list. I want to feel like I'm in the captain's chair with my 3 monitors, reclining while I oversee tasks complete themselves.

So my question to you all is - (1) what is the extent of your AI utilization today, (2) what does that look like in your day to day (here is where we can share tips and tricks on efficient use of AI), (3) how has that affected your productivity, and (4) where do you see your specific use of AI going from here?

153 Comments
 
Controversial

What is the extent of your AI utilization today? 

Zero. The end product is not high quality, it is terrible for the environment, and there are studies that show it is actively making people dumber (if not leading a small amount into genuine psychosis.) 

What does that look like in your day to day (here is where we can share tips and tricks on efficient use of AI)

My day to day generally looks like it always has, except for a higher amount of bullshit sent from LLM users. If someone sends me an email, or worse, a pitch, that is clearly written via LLM, I actively think less of them. If you can’t take the effort to write me something, I’m certainly not going to put effort into reading it. 


How has that affected your productivity?

It hasn’t. I don’t get paid based on productivity. I do not have a need to churn out machine slop. 

Where do you see your specific use of AI going from here? 

I’m not going to predict the future. Maybe LLMs revolutionize work or maybe it remains a big tech circle jerk. Let’s see what the landscape looks like post-bubble. 

Commercial Real Estate Developer
 

Gonna say it again...

AI IS NOT A BUBBLE.

Every new tech earnings report keeps proving you bubble alarmists wrong. Demand for this technology is insatiable and if you're not finding creative ways to use it, you will be left behind.

 

ojapar20

Gonna say it again...

AI IS NOT A BUBBLE.

Every new tech earnings report keeps proving you bubble alarmists wrong. Demand for this technology is insatiable and if you're not finding creative ways to use it, you will be left behind.

You saying something in all caps doesn’t make it any more true. 

“You will be left behind” is big tech propaganda. You don’t have to swallow the whole load. 

Commercial Real Estate Developer
 

Every new tech earnings report keeps proving you bubble alarmists wrong. 

And yet, none of those earnings reports actually give any detail about AI or LLMs.

You think that's maybe germane?

 

None of these companies are actually turning a profit on AI, and it's still uncertain if they'll ever turn meaningful profit. Almost every yc startup from this spring is a chatgpt wrapper or slop as a service, we are almost certainly in a bubble. I think the biggest problem is most of the easy gains from ai have been made. To get meaningful increases in efficiency from here would largely require people to put full faith in AI which almost no one would do.

 
Most Helpful

I agree with the sentiment of what you're saying, but nothing more. I also don't love AI creeping into every part of my life. It's coming for parts of my job. The algorithm on my phone pulls my attention away from stuff I actually find fulfilling in life. AI opens the door to ill intentioned mass surveillance. Etc.

But you can't let those feelings drive your career decisions. Treat your career like the investor you are. A prudent investor doesn't let emotion influence the facts in front of him.

And the facts in front of me: Copilot searches and manages my Outlook inbox instantly. I prompt it for commentary in the style and tone I want, then spend 10 min editing instead of an hour writing it all myself. I have it double check my work to find small error (especially valuable when you got no sleep). I have it find stuff in legal docs I can’t be bothered to read. And so much more that I find it difficult to believe you have put in a good faith effort to find use cases yourself.

Im willing to bet my career that AI is going to transform the corporation. I go out of my way to sit with our tech team and learn how they're building custom agents in Microsoft Studio for each LOB. Very soon they will put low/no code agent building in the hands of front office people. The ones who jump at the opportunity like their career depends on it will be the ones who get to stay in this industry.

CRE - You might be senior enough to ride this head in the sand stance out to retirement, if you’re lucky. Fine. But it’s a bet with zero upside if you’re right. Pushing it on a forum full of guys with 30 plus years of career left is irresponsible and WRONG advice.

 

sappy

CRE - You might be senior enough to ride this head in the sand stance out to retirement, if you’re lucky. Fine. But it’s a bet with zero upside if you’re right. Pushing it on a forum full of guys with 30 plus years of career left is irresponsible and WRONG advice.

Gen Z is the first generation that is dumber than the generation before it. Every single generation has outperformed the generation prior to it in cognitive capability…until Gen Z. Gen Z has lower scores in basic attention, to memory, to literacy, to numeracy, to executive function, and even general IQ. 

The challenge I make to a forum full of 20 year olds is: by not relying on LLMs to function, you will not be functionally retarded like the rest of your peers. And if I’m wrong, the barrier to entry for using LLMs is so low that anyone who drags their feet like I do will be “caught up” in a week. They won’t even have to catch up if AI is that advanced—AI will already be caught up for them. 

Commercial Real Estate Developer
 

CRE

What is the extent of your AI utilization today? 

Zero. The end product is not high quality, it is terrible for the environment, and there are studies that show it is actively making people dumber (if not leading a small amount into genuine psychosis.) 

What does that look like in your day to day (here is where we can share tips and tricks on efficient use of AI)

My day to day generally looks like it always has, except for a higher amount of bullshit sent from LLM users. If someone sends me an email, or worse, a pitch, that is clearly written via LLM, I actively think less of them. If you can’t take the effort to write me something, I’m certainly not going to put effort into reading it. 


How has that affected your productivity?

It hasn’t. I don’t get paid based on productivity. I do not have a need to churn out machine slop. 

Where do you see your specific use of AI going from here? 

I’m not going to predict the future. Maybe LLMs revolutionize work or maybe it remains a big tech circle jerk. Let’s see what the landscape looks like post-bubble. 

 I don’t think AI is "making people dumber" any more than calculators made people bad at math or GPS made people forget how to get around. Humans have always created tools to handle certain tasks so we can spend more time focusing on bigger ideas and more important thinking.

Like, writing didn’t make people less smart just because we stopped memorizing everything. Cars didn’t make us lazy because we walked less. Every generation has had new technology people were worried about, but usually it just changes how we do things, not whether we’re intelligent.

 

Like, writing didn’t make people less smart just because we stopped memorizing everything.

But it probably did make us worse at memorizing things, or listening for extended periods of time.  The point of writing is that it allows us to be lazy about that sort of thing.  You don't need to remember, if you can write it down and refer back to it later.  Whether that makes us "less smart" or not, it certainly qualifies as eroding a skill or ability.

Cars didn’t make us lazy because we walked less. 

Do you have evidence for this?  Intuitively it seems wrong.

Humans have always created tools to handle certain tasks so we can spend more time focusing on bigger ideas and more important thinking.

The problem with AI is that it doesn't let us focus on bigger ideas.  It's replacing that function for lots of people.  A calculator doesn't tell you what formula to use - you have to do all the work yourself, it just automates the calculation.  Want to figure out the weighted average sales tax in the United States?  If you want to figure that out with a calculator, you have to actually think about the problem.  How do you get to that, what inputs do you need?  With LLMs, you don't actually need to do that.  There is no thinking.  And the fact that you don't have to think about it means you aren't equipped to spot check the work that is being done.

 

mdwsteq

CRE

What is the extent of your AI utilization today? 

Zero. The end product is not high quality, it is terrible for the environment, and there are studies that show it is actively making people dumber (if not leading a small amount into genuine psychosis.) 

What does that look like in your day to day (here is where we can share tips and tricks on efficient use of AI)

My day to day generally looks like it always has, except for a higher amount of bullshit sent from LLM users. If someone sends me an email, or worse, a pitch, that is clearly written via LLM, I actively think less of them. If you can’t take the effort to write me something, I’m certainly not going to put effort into reading it. 


How has that affected your productivity?

It hasn’t. I don’t get paid based on productivity. I do not have a need to churn out machine slop. 

Where do you see your specific use of AI going from here? 

I’m not going to predict the future. Maybe LLMs revolutionize work or maybe it remains a big tech circle jerk. Let’s see what the landscape looks like post-bubble. 

 I don’t think AI is "making people dumber" any more than calculators made people bad at math or GPS made people forget how to get around. Humans have always created tools to handle certain tasks so we can spend more time focusing on bigger ideas and more important thinking.

Like, writing didn’t make people less smart just because we stopped memorizing everything. Cars didn’t make us lazy because we walked less. Every generation has had new technology people were worried about, but usually it just changes how we do things, not whether we’re intelligent.

It’s irrelevant whether or not you believe it. There’s plenty of data that shows it. I believe someone linked an example elsewhere in the thread. 


And car culture doesn’t make you lazy, but it is undeniably one of the primary reasons Americans are less healthy than people who live in walkable societies. 

Commercial Real Estate Developer
 

Dude AI is here and it’s the future. I bet you old architects were saying shit like this when Autocad came out.

And you know what they were probably right - architecture is basically a lost art at this point and we are worse off because of this technology in a lot of ways.

That doesn’t change the fact, and I stress fact at this point (I beleive you are not working for a company and are on your own so you probably are unaware) that everyone from the CEO to the analyst is using Claude, and it does save people a lot of time. For good bad or otherwise it’s here to stay.

Whether it’s a bubble or not I’ll leave to Michael Burry to pontificate, but I will say the internet was both revolutionary and a bubble.

 

CRE

What is the extent of your AI utilization today? 

Zero. The end product is not high quality, it is terrible for the environment, and there are studies that show it is actively making people dumber (if not leading a small amount into genuine psychosis.) 

What does that look like in your day to day (here is where we can share tips and tricks on efficient use of AI)

My day to day generally looks like it always has, except for a higher amount of bullshit sent from LLM users. If someone sends me an email, or worse, a pitch, that is clearly written via LLM, I actively think less of them. If you can’t take the effort to write me something, I’m certainly not going to put effort into reading it. 


How has that affected your productivity?

It hasn’t. I don’t get paid based on productivity. I do not have a need to churn out machine slop. 

Where do you see your specific use of AI going from here? 

I’m not going to predict the future. Maybe LLMs revolutionize work or maybe it remains a big tech circle jerk. Let’s see what the landscape looks like post-bubble. 

Bro you are only looking at what's out right now, not what will be out in a few years and the technological advancements that will come soon in AI. Until you know the cutting edge AI concepts and ideas that will change the game soon you're just wrong about it lol

 

Fair take and appreciate the data point. I will say that you're clearly senior enough to enjoy such a stance around AI. As the article mentioned, senior leaders have the highest conviction around not being personally displaced by AI. How I see it, since you're someone who's further in their CRE career, your value-add isn't spreadsheets or tasks, it's initiatives and relationships.

But for those who are mostly or exclusively measured based on productivity rather than driving results, I think we have to care about AI. Whether it's bad for the environment or a bubble, no doubt it's changing how work is being done. If the data isn't already supporting it, I truly believe entry level jobs are thinning because it takes fewer or none of these roles because of AI; the analyst can do what two analysts can do, or the associate can do associate-level work AND analyst-level work.

From where you stand, as long as you get what you need from your support team to run your deals at the management level, you don't need AI to write an email or pull data or create an Excel formula. Though I think you could still benefit, the benefits are probably marginal for you. Not so for most professionals junior to you.

Edit: And no I didn't use AI to write this lol (not flattering myself)

 

If you didn’t use a LLM to write that response, you’ve been using one so often that your natural writing sounds generated. 

That is a not a compliment. 

Commercial Real Estate Developer
 

Anyone who says they don't use AI in their day to day work will be out of a job in 3 - 5 years.  You are not productive enough as a human being to compete against people who are of roughly equal productivity who use AI to boost their productive output.  

Saying shit like "the results aren't good" is a fucking lame ass cope. Best of luck to you in the sunset of your career. 

 

Exactly. These people just don't get it.

There's actual hard data now showing how strong AI demand is. Revenues in cloud computing and semiconductors are literally going through the fucking roof, so I have no idea what data the bubble alarmists have at this point other than "muh, I don't like AI slop!"

@CRE @jl12  will be proven so horrendously wrong on this one. I can't wait.

 

People just assume this is 96 - 01 all over again, but they don't understand utilization at all.  Which is ironic when it comes from the RE professionals.  Is AI perfect? No, but is it all useless slop? No.  Every boom cycle has a bunch of grift and slop, that gets shaken out as success floats to the top.

 

ojapar20

Exactly. These people just don't get it.

There's actual hard data now showing how strong AI demand is. Revenues in cloud computing and semiconductors are literally going through the fucking roof, so I have no idea what data the bubble alarmists have at this point other than "muh, I don't like AI slop!"

OK, and how much are expenses going up?

Your constant focus on revenue is telling.  I guarantee you I could make billions of dollars selling Snickers bars.  I'll charge $100 per candy bar, and every sale comes wrapped in $500 worth of cash.  People will line up around the block to buy my product!

 

ioncewenttoschool

Anyone who says they don't use AI in their day to day work will be out of a job in 3 - 5 years.  You are not productive enough as a human being to compete against people who are of roughly equal productivity who use AI to boost their productive output.  

Saying shit like "the results aren't good" is a fucking lame ass cope. Best of luck to you in the sunset of your career. 

It’s funny how emotional AI evangelists get about their little predictive chatbots. 

Commercial Real Estate Developer
 

Emotional? No, the only possible way to be emotional about this is if you are in an anti AI position. Understanding that AI is going to replace you is just a sober acceptance of reality. Learning how to adapt to that is how you will thrive. 

 

ioncewenttoschool

Anyone who says they don't use AI in their day to day work will be out of a job in 3 - 5 years.  You are not productive enough as a human being to compete against people who are of roughly equal productivity who use AI to boost their productive output.  

Saying shit like "the results aren't good" is a fucking lame ass cope. Best of luck to you in the sunset of your career. 

What, specifically, does AI do for you?  How does it make you more productive, and by how much?  What are you doing now that you couldn't or didn't do 10 years ago?

The results do suck.

 

I have litterally built a database to process and analyze 40 years of offering memorandums to track historical trends across MSAs that the portfolio operates in. Not only do we track real estate trends with this data analytics tool, we also track business health in these areas because we also purchase lower middle market to middle market businesses.  

This is litterally something we had zero capability to do prior to this.  But yes, AI is just fluff for your wife to vent to about you. 

 

They don't have access to or aren't using the latest models.  AI has gone god mode these past 6 months.  The technology is amazing, its the figuring out the exact use cases, and path to profitability for enterprise AI, that's left now.  But as a tool, fucking mindblown.

 

Daemon145

They don't have access to or aren't using the latest models.  AI has gone god mode these past 6 months.  The technology is amazing, it’s the figuring out the exact use cases, and path to profitability for enterprise AI, that's left now.  But as a tool, fucking mindblown.

Any examples?

...but is it REPE?
 

ioncewenttoschool

The AI haters really get their panties caught in their ass about this.

You’re the only person I’ve seen in the thread, including me and the OP, who is being emotional about this. 

Maybe ask chat for a lil pep talk. 

Commercial Real Estate Developer
 

On to my thoughts of the article specifically.  It is cope. While it is true that RE is less exposed to AI than other areas, it is insane to think that it is somehow magically safe. As someone who has built companies in CRE, Proptech, RE AI, and Finance I can tell you that the industry's understanding of the ability of AI to disrupt the industry is significantly understated.

 

AI saves me probably 10+ hours a week when I am making fundraising decks. It totally changes the game and makes my slides look a lot better and require me to think less.

Emails too, I used to spend an hour crafting some emails to send to the top of the chain. Now I can spam thoughts and have it make something that I can continue to edit. Turns an hour email into 15-20 mins.

Underwriting, it is kinda useless for me at this point. I can probably break a MF rent roll down and know it’s accurate by the time I upload it to the AI and ask if some questions. AI could also mess things up too.

 

Associate 3 in RE - Comm

and require me to think less.

And there lies the problem 

Commercial Real Estate Developer
 

Associate 3 in RE - Comm

AI saves me probably 10+ hours a week when I am making fundraising decks. It totally changes the game and makes my slides look a lot better and require me to think less.

Emails too, I used to spend an hour crafting some emails to send to the top of the chain. Now I can spam thoughts and have it make something that I can continue to edit. Turns an hour email into 15-20 mins.

Underwriting, it is kinda useless for me at this point. I can probably break a MF rent roll down and know it’s accurate by the time I upload it to the AI and ask if some questions. AI could also mess things up too.

How the fuck did it take you an hour to write an email? 

Good lord you deserve to be replaced by a machine. 

...but is it REPE?
 

If it takes you an entire hour to compose a grammatically correct email, then maybe you aren't the best example to be considering here.

 

From reading this thread and other comments about AI in the industry, it seems like many still think we’re in the LLM chatbot phase. In reality, AI has advanced to the point where the average joe can create AI agents to do anything they want, ranging from having the agent provide daily updates about internal workflow and decision making monitoring if you’re managing others, to coding software that can help the company with deal screenings, asset management or portfolio management duties. If I’m a property manager I can have an agent bid out a project to vendors, filter the bids and provide me a summary to choose from. That’s an example of time and energy saved.

 

Associate 1 in RE - Comm

Stop saving time and energy!!! You're supposed to spend MORE of it on MENIAL tasks so that your brain doesn't instantly wither away and now you're in a VEGETATIVE state because ARTIFICIAL INTELLIGENCE

“Why waste TIME talking to women?! It is so much more EFFICIENT to DATE my CHATBOT! I named her SCARLETT and she tells me I’m a GOOD BOY!”


Man, why are AI bros so fucking lame? 

Commercial Real Estate Developer
 

ioncewenttoschool

Don't worry the old Millennials and young Gen Xer's will lobby the government to seize more of the value you create with AI to fund their head in the sand world view.  

You DEFINITELY have an AI girlfriend 

Commercial Real Estate Developer
 

Exactly.

The people calling AI a bubble simply haven't used the technology enough to understand its use cases. I coded a side project in about 2 weeks that's generated me about $1000/mo. Without AI, this same website would've taken me months to develop. Ordinary people can now build whatever they want. This is huge.

 

ojapar20

Exactly.

The people calling AI a bubble simply haven't used the technology enough to understand its use cases. I coded a side project in about 2 weeks that's generated me about $1000/mo. Without AI, this same website would've taken me months to develop. Ordinary people can now build whatever they want. This is huge.

I genuinely don’t know why I’m supposed to be impressed by this. “Ordinary people” could always build whatever they want. They learn the skills and then do it. 

You didn’t learn the skills. You didn’t do it. Congrats. You make…another $12k a year. What a revolution. In four years, you can buy yourself a Camry. 

Commercial Real Estate Developer
 

I've been using it to build tools that automate a bunch of accounting work. Also use it to respond to people who obviously use AI to write an email since I'm not going to put effort into responding to it.

I've been looking into AI software a lot recently and it seems like CRE has gotten past the worthless chatbot phase and is going to focus more on report creation through prompts (for the multifamily side at least). Let's say you want to find any leases that have a loss to lease greater than $75 across your whole portfolio. Instead of running a bunch of rent rolls or GPR reports, exporting them and chopping them up to get the usable information, you will do a prompt like "look across our portfolio and return any units with a LTL greater than $75". 

So I think the next big push will be that and analysis. There were some brief mentions of it doing variance analysis which I think is the wrong way to go since you use that as a way to get the managers to tell on themselves.

 

One topic the article covers is the extent CRE professionals have woven AI into their work. Apparently 76% use it as a basic assistant, but only 5% use it to automate their workflow. 

This sounds pretty accurate to me.  Lots of bosses are forcing AI functionality down the throats of their employees, but employees don't want to use a shitty product that churns out endless slop and doesn't actually make their lives meaningfully easier.  It's telling that we never really get any information about what it means to use AI as a "basic assistant."  What kind of engagement does that mean?  If I google something, I get an automated AI output at the top of my search results.  Does that count as "use" even if I ignore it?

It increases my productivity in maybe 50-60% of areas, sometimes in big ways (e.g. summarization, drafting emails, market research, Excel templates from scratch) and often in only marginal ways (e.g. regular reporting, legal doc analysis, CAM Rec review).

I feel like this overstates the case.  50-60% of the tasks you do might only constitute 5-6% of the time and brain power you expend in a given day.  How many Excel templates do you need to build from scratch?  How long as "drafting an email" take?  If you function as a mindless robot, then yes, I can see how AI will automate away your job.

All in all it's been hard to really integrate it into my workflow but I feel like it could free me up by 70%+ if I could just have it understand my ask and pointed at the data I need. 

Yes, if the chatbot was capable of doing the job of a human, it would be amazing.  But it can't.  That's the point.  AI doesn't think.  It simply scrapes lots of data and makes an educated guess about the next link in the chain of reasoning.

(1) what is the extent of your AI utilization today

Effectively zero, and I don't see that changing in the near future.

(2) what does that look like in your day to day (here is where we can share tips and tricks on efficient use of AI)

To the extent I use it at all, it's to save me a couple clicks in solving basic facts (e.g. the population of a given neighborhood or city when writing an IC memo).

(3) how has that affected your productivity

Given how much time I have to spend clicking out of prompts from the various AI functionalities being forced into my software, it probably nets out to no impact on productivity.  The 5 minutes a week or so I save using AI is offset by that.

(4) where do you see your specific use of AI going from here?

Probably nowhere?  I can see some uses for technology (not specifically AI) in my life, but few of them are within my remit to implement.  I have some interesting ideas for using pattern-recognition capabilities, but nothing transformative and nothing that requires "AI"

 

larrysellers

You're safe Ozy, 4 unit brownstones in brooklyn can still be u/w on the back of a napkin. No AI needed.

Your awful attempt to belittle me notwithstanding, you've inadvertently brought up another reason AI is useless.

A chatbot isn't going to tell you what is in the walls of that Brooklyn brownstone.  When you or I or anyone else wants to do that rehab job, the only important thing is how much that work is gonna cost.  And the only way to figure that out is to have a human being go and look at the conditions and estimate the budget.

 

From the comments, I could see both being true.  

- AI use increasing exponentially and impacting more repetitive work or technical work, thus replacing a lot of people. Customer service type work, for example.  Or, website development (huge impact). 

- AI use reducing some aspects of higher finance work, and at larger organizations, reducing headcount as a smaller percentage, or creating efficiencies. But at a slower pace. 


I don’t think you should ignore how impactful AI will be, and other technologies that are coming.



“My grandmother born in 1890 saw the world at a top speed of 40 miles per hour.  When she was 13, the Wright brothers flew an airplane for the first time... in her 30’s, commercial radio started… in her 50’s, commercial television started… and she lived long enough to watch a man walk on the Moon.”
 

We will see dramatic change in our lifetimes.  Maybe less so to the physical building world, but plenty in technology.

 

Have compassion as well as ambition and you’ll go far in life. I am interested in digital immortality. Check out my blog at digitalimmortality.com
 

As much as the broccoli-haired analysts in this thread want to paint me as some sort of Luddite, I completely agree with "We will see dramatic change in our lifetimes.  Maybe less so to the physical building world, but plenty in technology." Hell, we already have! 

I do not think LLMs, however, are part of that change—the same way crypto wasn't, the same way NFTs weren't. Not every new technology needs to be evangelized, and I would argue that the more a technology is evangelized, the more likely it is to be utterly disappointing. 

Even calling it "AI" is an intentional misnomer driven by fundraisers and marketers. LLMs are not "AI." Real AI would be transformative. Possibly ruinous! We won't know. What we do know is that Grok, ChatGPT, Gemini, Copilot, etc. are anything but transformative, and are primarily lauded by people who stand to make money from their success (tech bros, executives looking to cut staff), morons ("It took me an hour to write an email"), or talentless hacks (AI "artists" and "musicians"). 

And that's ignoring any moral argument against it, which is AI datacenters using more water than the bottled water industry and polluting communities, mass surveillance by companies like Palantir and Flock, people becoming literally dumber by using it, the entire technology being built by of stealing work and intellectual property, etc.  

Commercial Real Estate Developer
 

Jesus fucking Christ. Comparing LLMs to crypto is just stupid af.

AI is automating 6 FIGURE JOBS in software engineering. And I strongly believe this is because the people at OpenAI and Anthropic are also software engineers themselves, so this is the first thing they'll know how to automate.

Eventually, this will trickle to other industries. I'm not saying there will be literally ZERO jobs, but reducing LLMs to AI slop and email drafting is ridiculously shortsighted.

I hope you revisit your comments in 10 years. You will be proven hilariously wrong.

 

Associate 1 in RE - Comm

We can close this thread. Odog has balanced the scales of truth, and the nonchalant Luddite CRE gets the short stick.

SO emotional about this 

Commercial Real Estate Developer
 

When I am doing code shit it is marginally useful, but it is also producing client emails where they feed something into an AI and then ask questions that could be answered in the document that they fed into the AI. Also the AI will straight up make shit up. Where in this document do you see a prior sales price of $X? Your AI can say something, but then a human being has to waste time telling you why your jumped-up Markov chain is making shit up.

 

I use it pretty frequently for things like legal doc abstraction, OM summaries, etc.  I have also found the Excel Claude plug-in to be very useful for throwing together simple calculations for taxes and equity waterfalls.  A lot of these tasks are things that I previously would've passed down to an analyst, so I can see some pressure on employment there.  

Overall, it probably saves 5-10 hrs per week.  I am in the camp that it is both useful and overhyped.  I think eventually it gets profoundly more expensive than it is now when more of the costs get passed down to the end user.

I will say one of our executives has AI psychosis and is constantly sending us slop from Co-Pilot as a substitute for reading source material and independent critical analysis.  It drives me absolutely nuts. 

 

Ex MBB here, working on AI in Tech now. For generic work, AI is already useful: summarizing docs, turning messy notes into a memo, drafting first-pass emails, cleaning up language etc.

But the bigger shift is agentic workflows: systems that can take a goal, pull from multiple sources, complete a sequence of tasks, flag exceptions, and hand a mostly finished work product back to a human. The issue is that most CRE firms are nowhere near ready for this. Their useful data is buried in PDFs, email threads, OneNote notes, Yardi/Argus exports, old IC memos, loan docs, and random Excel tabs.

So I think both sides are partly right. CRE is less exposed than pure screen-based jobs because physical assets, relationships, local judgment, diligence, and capital decisions still matter. But juniors whose main value is processing documents, formatting decks, updating reports, and moving information between systems are absolutely exposed.

Read more of my writing here: https://consulting2tech.substack.com/
 

Anthropic has itself admitted that they've seen no layoffs related to their products.  The company with the most to gain from touting the ability of their software to automate jobs is basically confessing that their marketing has been a giant pile of bullshit.

Can we please stop assuming that what AI companies claim is true, and start assuming the opposite?  Every single one of these labs has a long history of lying through their teeth about capabilities, about what the next iteration looks like, about basically everything. And yet this thread is perfect evidence of the fact that there are still tons of people inclined to give them the benefit of the doubt.  The default position should be "prove it" and yet even after we've seen the rug pull time and time again, you've still got AI fanboys and boosters willing to gobble up this shit sandwich and praise the taste.

 

I have used Claude Code and Claude in Excel to build models and do heavy duty analysis for me over this past month (eg cutting up 250k row bookings databases). I am not using the free version of Claude btw, I am using the top end models and basically have unlimited tokens (last month I burned ~$1k in tokens) so I am maxing out the capabilities. It is very good and I've been able to get a project done by myself that would've 100% required a junior analyst last year. 

'Ex-consultant in Tech' actually had a balanced comment. A lot of the responsibilities at the principal level of CRE will not get automated - walk throughs, broker / contractor / capital provider relationships, judgement, salesmanship etc. But the analyst level number crunching 100% will be in the next ~2 years based on how impressive the agentic tools already are.

Anthropic's CEO Dario is the one going around telling everyone that all white collar jobs are about to disappear.. where are you seeing they confirm no layoffs from their products? On that front, I think he really needs to dial it back or he risks pissing off the population even more than he has. 

 

Interesting thread. I’m seeing a similar gap — lots of people using AI for drafting and summarizing, but far fewer actually redesigning workflows around it. The real unlock feels less like “better prompts” and more like connecting AI to the right data and repeatable processes.

 

I think this thread makes clear one thing:


Despite being profilic WSO posters, CRE and Ozymandia clearly do not work for large scale institutional firms. For younger folks wanting to work at those firms, do not listen to them 


Here’s why: all of the major multi-asset class private equity firms are actively integrating AI across their organizations


They are encouraging their employees to use AI often


While yes some of these firms are also invested in AI, this is not a conflict of interest.


in fact that opposite, it’s an unbiased realization from putting great minds in the same room, debating AI, and realizing this is a revolutionary technology 


If you want to work at a real firm, use AI, learn AI, take a view on how AI will impact the economy and real estate.

Do not listen to these fools 

 

It’s interesting how the institutionalization of commercial real estate over the past two decades has led people to denigrate entrepreneurs in favor of corporate drones at “real firms.” I suppose it was inevitable, but it’s kind of sad at the same time. 

Game’s gone. 

...but is it REPE?
 
[Comment removed by mod team]
 
[Comment removed by mod team]
 

Nostrum error et pariatur ducimus. Et molestiae maiores cupiditate ea cupiditate occaecati et. Aut est in illum dolorem et.

Et sunt aliquid maiores est fuga et eos. Optio nemo velit iste consectetur quis rerum. Ipsa magnam tempora doloribus adipisci recusandae. Dolorem quibusdam cupiditate adipisci recusandae.

Cupiditate repellendus porro cupiditate voluptates est impedit. Velit eius cupiditate consequatur laborum velit iure. Molestiae enim facilis quaerat eveniet officia.

Numquam architecto aliquid hic enim aperiam numquam ut voluptates. Perspiciatis est enim distinctio deleniti et. Et natus id adipisci ea rem omnis.

 

Repellendus exercitationem explicabo qui. Incidunt veniam hic ex nam aut mollitia. Omnis ullam enim labore saepe aperiam repellat delectus voluptas.

Mollitia autem repellendus voluptas esse excepturi consectetur aliquid. Sapiente consequatur praesentium enim ea. Voluptatem id veritatis debitis ratione molestiae qui dolorem. Mollitia iure natus est iusto voluptate et. Inventore quaerat ea occaecati enim sapiente. Eum facere aut magni.

 

Maxime quod recusandae quia voluptates laborum illum. Et ex tempore quaerat qui omnis esse odit. Reprehenderit rerum labore non reprehenderit unde. Aperiam blanditiis enim fuga incidunt odio.

Numquam occaecati inventore illum sit iure cupiditate aut. Et nemo necessitatibus ullam laboriosam voluptates. Quo eum et voluptatem ut. Consequatur at eos vel quae.

Qui quia quae necessitatibus non quo. Amet aut ipsa ab sit ut. Quam id voluptas dolor quae adipisci aut. Tempora rem alias alias porro unde soluta eveniet. Ipsam tempore odio beatae rerum eos ea iure. Dolorem voluptas et nihil atque. Odit aperiam ut ut aut qui nisi non aliquam.

Magni libero voluptates eum autem asperiores. Voluptatem nihil perspiciatis ullam culpa accusantium. Ratione vel dolore laboriosam quaerat.

 

Tempore placeat velit in voluptatem quas illo. Ipsam maxime voluptatibus aliquam aut dolorem. Omnis fugiat eligendi temporibus ab quia. Saepe beatae doloremque aliquam et.

Repudiandae nostrum ut veritatis quidem optio. Voluptatibus molestiae magni rem sed. Non asperiores natus velit nobis omnis dolores.

Career Advancement Opportunities

July 2026 Investment Banking

  • Evercore 01 99.4%
  • Moelis & Company 01 98.9%
  • JPMorgan 01 98.3%
  • Guggenheim Partners 01 97.7%
  • Morgan Stanley 07 97.1%

Overall Employee Satisfaction

July 2026 Investment Banking

  • Moelis & Company No 99.4%
  • Morgan Stanley 02 98.8%
  • Evercore 01 98.3%
  • BMO Capital Markets 12 97.7%
  • Banco Santander 01 97.1%

Professional Growth Opportunities

July 2026 Investment Banking

  • Evercore 01 99.4%
  • Moelis & Company 01 98.9%
  • Morgan Stanley 06 98.3%
  • Goldman Sachs 01 97.7%
  • JPMorgan No 97.1%

Total Avg Compensation

July 2026 Investment Banking

  • Vice President (15) $434
  • Associates (45) $258
  • 3rd+ Year Analyst (8) $210
  • 2nd Year Analyst (22) $179
  • Intern/Summer Associate (13) $156
  • 1st Year Analyst (79) $150
  • Intern/Summer Analyst (73) $101
notes
16 IB Interviews Notes

“... there’s no excuse to not take advantage of the resources out there available to you. Best value for your $ are the...”

Leaderboard

1
redever's picture
redever
99.2
2
kanon's picture
kanon
99.0
3
BankonBanking's picture
BankonBanking
99.0
4
Secyh62's picture
Secyh62
99.0
5
DrApeman's picture
DrApeman
98.9
6
GameTheory's picture
GameTheory
98.9
7
dosk17's picture
dosk17
98.9
8
Betsy Massar's picture
Betsy Massar
98.9
9
CompBanker's picture
CompBanker
98.9
10
bolo up's picture
bolo up
98.8
success
From 10 rejections to 1 dream investment banking internship

“... I believe it was the single biggest reason why I ended up with an offer...”