I was listening to a news story today about "AI Agentic".  In simple terms, it's a way of using AI to do basic--and sometimes complex--research and analysis.  One might ask AI bots to go off--essentially on their own--to gather and analyze data, and report back, on the top news stories of the day, based on whatever parameters you set.  The descriptions emphasized that these basic "busywork" tasks and basic analytical tasks that are often those that used to be done by humans.

 
Which brings me to one of our favorite topics on this board:  Vetting.  Right now, both players and providers spend a great deal of time making sure the other is someone they want to see.  If we could automate that process, conceivably we would save ourselves a lot of time.  And if the AI Agent is well-enough trained, possibly it could provide a daily or weekly list of the providers we want to see based on multiple criteria; and produce a list of players a lady would be willing to entertain,  based on her criteria.  Not sure about the wisdom of letting AI bots loose on the data in TER, provider ads, OF, RW data, etc., all in service of getting laid, but I gotta believe there is some clever person out there giving this serious thought right now.

I think they would be among the first to look at incorporating AI into their business model.

AI might eventually play a role in this lifestyle, but there are already plenty of unsettling stories circulating about AI in general, some even coming from its own creators. lol. Jokes aside, the concern does seem to be growing. Still, it’s possible AI could be useful for p4p in some capacity down the line, though the illegality of our litle hobby might prevent algorithms from enabling those kinds of requessts.
.
I’m also not sure how we would ever configure it to return matches based on physical attraction, since that’s such a subjective and unpredictable thing. Yes, I have a general “type” I am atttrated to but I don’t stick to it rigidly, and even within that type, attraction can vary quite a bit. How would AI help there?
.
What I don’t understand is why existing platforms don’t offer better, and more, filtering tools. For example, on a site like Tryst, you can filter by things like body type, age, cost, style, etc....but why not also filter a specifc provider in a specific city on specific dates? That would be a game changer!
.
Take this scenario...I’m traveling to LA for work and have three evenings free. If I’m looking for p4p during that window, I currently have to click through individual profiles just to figure out who’s active, who’s touring, and who’s even available. Then I have to check pricing, screening requirements, deposit demands (both for type AND amount), and after all that, still verify and assess compatibility and then hope the girl gets back to me in a reasonable time frame. It’s exhausting just thinking about it. lol.
.
At that point, a well designed database that could give results meeting almost all those criteria would be more than enough and then I probably wouldn’t even need AI. I mean, a man can dream cant he? lol.

Very few tap into the full potential of AI. It's rapidly growing. Many are being slow to testing the waters. And, many who are, are using programs like ChatGPT, Gemini and Claude. And, we're not even using them to their potential. I'm included in this group. But there are so many more programs and so much more they can do then many of us realize.  

 
A large part of what AI does for you is both based on the program you're using the prompts you give it. For someone savvy enough, giving AI access to your TER account is fairly simple. The more difficult part is what prompts (instructions) you will give it. AI is not going to figure this all out on its own. If someone with skill wants to build out a program and share/sell it to others, they can. And, this is the case for anything on a website. AI can easily scrape data and follow whatever instructions you provide it. For all we know, someone might have tried it already.

 
This brings us to the next challenge. Will it actually produce what you want? If in design there are flaws, you might not even realize this to be the case. With anything new, you can do all the testing in the world, and you still might discover things later. At this stage, AI is best used as a tool that has human oversight in cross checking. It can save you a ton of time. But if you get lazy about it, it can also be problematic.

 
A lot of AI sources data from past decisions. If your past decisions had flaws, it will learn to repeat those flaws if you don't cross check, catch them, and provide prompts to correct or filter them. I do think if done correctly it can be useful. I'm always a fan of making things more efficient and saving us time.

 
Interesting topic. I'm glad you're bringing it up.

-- Modified on 4/22/2026 12:52:18 PM

Do you want AI agents to read reviews for you and make recommendations based on some criteria?

 
If so, I'll remind people ter has rate limits on review viewing. As someone who actually has scraped data off the site (not with Ai, I have multiple cs degrees), I will tell you - first they have captcha set up after reading a certain number of reviews. So your Ai will have to solve a captcha. Now, solving it in today's world isn't as diggivult as it used to be. But still.  

 
Then - there is an alert sent to ter if someone had viewed too many reviews. I believe it's 300 over two days (will need to check my notes/code). Then an alert is generated and people investigate. Hence I was always careful to not hit that limit.

Ai needs a lot of training data to pin down your criteria. I'm just not sure how it will help in a way you yourself can't.

I assume the goal is to save time. Obviously, there is an initial time investment in building this out. Makes less sense if everyone is building it out there own. But if someone builds something out that others use, then the only person investing that part of the time is the builder.  

 
TER's filtering would weed out ones that there is no reason to scrape the reviews, so all of those would factor out of the quantity concern. And 300 in two days is A LOT. By the time I enter details like city, age, price range, BBBJ, etc... that TER allows, the number gets smaller. Also, probably no value in scraping those with no recent reviews.

 
I just tried some filters and only got 66 matches in Los Angeles. I expanded it by removing the city, it's now 770 matches. This includes those without recent reviews, so in both cases numbers would shrink. Now, the part this doesn't tell is how many reviews would need to be read for each provider, as that falls into your limits point. For that reason, it won't work as fast as some might like. But it's still more than most of us would do. If we are talking reading the full reviews, personally I'm not reading even 20 in a day. I usually skim and read more details if enough catches my eye. Even then, I read a few reviews initially. I only dig deeper if I decide I want to book her.

 
Anyway, I think it's an interesting idea. But it's far from simple, and you bring up good points about barriers I was not aware existed.

My question was what exactly to automate. I know that many people view Ai as a time saving tool. I just don't really see what it offers here if you can filter search yourself and it takes seconds. You will spend more time training it what your preferences are as well as criteria, than the time it will save imo

I would think it would be filtering out certain details that show in the reviews that might not show in their profiles, and therefore not be filtered out using TER filters. I have seen plenty of profiles say kissing with tongue, but found in their reviews it was more LFK. I have seen profiles say blow job without condom, but then in reviews someone points out it’s CBJ. I have seen profiles say natural boobs, but man made in reviews. So, using review details to cross check these details.

 
Depending on the person, there might be other things that matter. TS reviews don’t offer a filter that clarifies if they top, bottom or are verse. Can they get hard/stay hard? Do they cum? But these often show in their reviews. Those seeking BDSM might have certain fetishes they want. I imagine if you read reviews, other turn offs and turn ons might jump out. Maybe you want a squirter? Maybe you want to filter out reviews where the provider is mechanical? Maybe you feel like anyone with less than ten reviews isn’t reliable? Maybe you feel like anyone with over 250 reviews is too high volume?

-- Modified on 4/22/2026 10:28:14 PM

Yes, someone or AI bot will come up with an algorithm..... the only caveat is .....🤔

I would think you can do that with a much simpler bot or simple python script. I think rockets point about people not really getting what AI and the agentic versions of them can and cannot do. But yes, you could probably do some vibe coding to get what you want done.

Now, I will admit, I’m not the most knowledgeable about where bots/scripts become less applicable and AI would be better. That said, my examples were meant to touch the surface. Often, when I cross check I am also looking at ads, socials, and what might show up on the web. I don’t do this last part per se, but if AI could also cross check board posts that would be swell too.

…for lifelike ai sexbots to become mainstream…i suppose an unholy alliance of sex workers, religious extremists, feminists, and anti-technology activists could precent it from happening

regarding your post meta, nike, and microsoft are laying off people thanks to advances in ai…it is becoming more and more robust by the day where menial clerical tasks can be done with a click and they only really need a few highly
skilled person to check the ai output

We all have our "My Favorite" filters... Two of the filters that are important for me ...
1) Deposit Required,  No deposit required, Deposit at my discretion (as possible check boxes)
2) Dates when the provider is visiting  

Other filters would be the specific services, if the ladies are willing to list them.

My take is that it comes down to having something that will search and make decisions on its own or just return some search results based off defined criteria. Both cases are what a person does, but it takes time and effort. But my take is that any AI, even after a lot of training, is not going to make many really good decisions for you and you end up having to filter through the AI recommendations just as you need to do with a bot/script/filter search (the filters are site based so bots/scripts help with consolidating into a common view of the data for you).  

 
What people are reading about all the agentic functionality and how it does tasks as well as or better than humans is about work tasks that some company has spend millions developing and training. None are well trained to search for escorts so you're going to do that. How? Using some of the cloud services will be tricky I think. Doing it "in house" is possible but you'll need to know how, and might need to put some money into the system that is running the software. Having a site crawling bot is much simpler to make and run and you probably don't spend much more time reviewing the results.

 
All that get compared to just how much time and effort you put into a search for your next appointment. If you're only spending half and hour to an hour as week or so not sure there is much bang for the buck with either approach.

Yes, I do think AI can do the analysis. Using the contents of my reviews; my discussion contributions; the ladies I have seen (their attributes); other reviews; etc...
For example, in my replies to the General Discussion Board, I know I have mentioned number of times I will not make deposits and AI should use that information in the analytics...

Yes, I do think AI can do the analysis. Using the contents of my reviews; my discussion contributions; the ladies I have seen (their attributes); other reviews; etc...
For example, in my replies to the General Discussion Board, I know I have mentioned number of times I will not make deposits and AI should use that information in the analytics...

Yeah, no chance I have the skills or resources to create something like this. I would use it if someone else developed it. I might consider paying to use it. My time is worth money. I do think the difference here is you’re rationalizing what can happen now. I’m talking more what could be someday. AI will grow, it will become easier to expand, and it will require less resources to do so. It’s inevitable.

When computers were in their infancy, they were expensive and had limited functionality. Today, a savvy hacker can screw up public infrastructure or extort a major corporation by hijacking their own data, you can find all manner of illicit computing tools on the dark web, and your cellphone has more computing power than the first few generations of desktops. Most of humanity didn't conceive of the idea of machine learning (what they used to call AI) until a couple of years ago (at least outside of science fiction) and look where we are now. Imagine where we'll be in another few years. And, remember when sophisticated tools such as GPS were the exclusive, cutting-edge province of the military?

Sure, we may not soon develop cheap and easy software to find the perfect date. That was never my point. But an agent that could quickly sort through prices, reviews, board posts (both here and elsewhere), ad copy, and yes--some day photos, might be cheap and easy enough that we'll adopt it like the other 50 apps we each have on our phones. Not as a substitute for final decision-making, but as a handy funneling tool.

-- Modified on 4/25/2026 7:16:30 AM