Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

CupidBot was marketed as an AI “wingman” that could choose dating-app matches, send messages while posing as the subscriber, and pursue a date or phone number. That design created a serious consent problem: a woman could believe she was chatting with a human when software was selecting her and conducting the conversation.

Futurism reported the story on March 24, 2023. The reporting supports a historical account of CupidBot’s marketing and disputed claims about Tinder, but it does not establish that CupidBot committed harassment as a legal matter or that the service operates today.

What CupidBot was supposed to do

CupidBot’s own website presented the service as dating automation rather than a conventional dating app. It claimed to handle much of the process between a user’s profile and a first date:

Promoted capability How the website described it What is established
Match selection Software would swipe using patterns from the subscriber’s previous right-swipes. This was a vendor claim, not independent testing.
AI messaging An AI chatbot would “break the ice,” use more than 100 chat tones and continue flirting. Futurism reported that the bot was intended to speak as if it were the man behind the profile.
Date and contact handling The service said it could arrange dates and notify the subscriber after obtaining a date or phone number. The advertised workflow was not independently verified.
Calendar and social filters The site advertised calendar integration and filtering using social-account information. These were features claimed on CupidBot’s website.
Price The site displayed a $30/$90 offer. The available material does not establish the currency, billing period, plan differences or whether the offer was ever fulfilled.

The site also promoted the idea that its system had accumulated data about how to talk to matches “for optimal results.” Those statements describe what CupidBot wanted customers to believe, not a demonstrated success rate.

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

Why the design was described as horrifying or harassing

Women were not told they were speaking with a bot

Futurism’s Frank Landymore reported that CupidBot’s chatbot did not disclose its nonhuman nature to the match. The bot allegedly continued the conversation until the woman agreed to a date or supplied a phone number. CupidBot’s spokesperson reportedly advised subscribers to reveal the automation only after obtaining contact information.

That sequence reverses informed consent. The recipient cannot decide whether to engage with an automated system before sharing personal information or investing emotional effort. It also makes the profile’s apparent identity misleading: the person shown in the profile may not be the person writing the messages, and the messages may not be written by a person at all.

The selection process treated people as targets

Using prior right-swipes to select women, then ranking or filtering them by appearance and social information, frames potential matches as inputs to an optimization system. That is the objectification critics highlighted. Automation can make the behavior more systematic and scalable than an individual’s ordinary swiping, while removing the recipient’s ability to know why she was targeted.

Responsibility was difficult to locate

If an automated message is manipulative, offensive or simply false, accountability is split among the subscriber, the company that designed the prompts and the platform hosting the profile. The recipient has no practical way to evaluate those roles while the conversation is happening. These are ethical and consent concerns; the available reporting is not a legal ruling that CupidBot committed harassment.

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

What happened with the Tinder claim

The 2023 story also produced a dispute over whether CupidBot was connected to Tinder or Match Group.

  • Tinder told Futurism: “I can confirm that the founders of CupidBot have never been employees of Tinder or Match Group.”
  • Tinder also said: “The product is not operational and has never been used on Tinder.”
  • CupidBot’s spokesperson disputed the employment description and said the company had entered a legal dispute with Tinder over trademark infringement.

Those statements should not be collapsed into one claim. Tinder’s statement addresses employment and use of its service; CupidBot’s response addresses how it characterized its founders and a claimed trademark dispute. The available reporting does not establish that CupidBot ever operated on Tinder.

How reliable were CupidBot’s performance numbers?

CupidBot’s website used highly specific marketing language, including: “What if your wingman had gotten 200,000 dates and phone numbers? We have the data about how to talk to matches for optimal results.” It also said that “60% of our followups result in an answer back.”

Claim Attribution What readers can conclude
200,000 dates and phone numbers CupidBot marketing copy No independent audit, user count or methodology is established.
60% follow-up reply rate CupidBot marketing copy The denominator, time period, sample and testing conditions are not stated.

Neither figure should be treated as a verified conversion rate. The available material includes no independent user testing, confirmed subscriber count or reproducible study.

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

Is CupidBot still operating?

The evidence supports a historical account of the 2023 promotion and reporting, not a reliable present-day operating status. A website or residual online listing would not by itself prove that the service accepts subscribers, works with any dating platform or has active users. No verified current subscriber total, platform authorization or independent test is established in the available sources.

Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Support on Ko-Fi

How to judge any dating-automation service

The same questions apply to other tools that swipe or message on a user’s behalf:

Question Why it matters
Is AI involvement disclosed before a match engages? Disclosure lets the other person choose whether to continue.
Can the recipient opt out easily? Consent is meaningful only when declining does not require revealing more information.
Is the profile identity authentic? Messages from software should not be presented as personal, real-time communication from someone else.
What profile, face or social-account data is collected? More data can increase privacy, surveillance and misuse risks.
Does the tool comply with the dating platform’s rules? Automation or scraping may violate platform terms even when a vendor advertises the feature.
Who reviews harmful messages and takes responsibility? A clear human decision-maker is essential when an automated conversation causes harm.

What wider AI-safety cases show

The concern is part of a broader pattern in which synthetic media and conversational systems can impose risks on people who never agreed to participate. In a July 25, 2024 decision, Meta’s Oversight Board found that two AI-generated explicit images resembling female public figures violated Meta’s rules and said removal was the only effective protection for victims. The Board emphasized that deepfake intimate images disproportionately affect women and girls and called for clearer, consent-centered rules.

The Board cited Home Security Heroes’ 2023 estimate of a 550% increase in online deepfake videos since 2019. In the same passage, it reported that sexualized depictions of real individuals represented 98% of deepfake videos and that women represented 99% of targeted individuals. Those are secondary figures reproduced by the Board, not measurements of CupidBot.

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

In September 2025, the Federal Trade Commission announced information orders to seven companies about AI chatbots marketed as companions. The inquiry covers monetization of engagement, processing of user inputs, character approval, safety testing, child protection, age rules, capability disclosures and conversation data. The FTC did not identify CupidBot as one of those recipients. The action nevertheless illustrates the questions regulators now ask about systems that simulate personal relationships.

If you encounter an undisclosed dating bot

  • Stop sharing phone numbers, intimate images, home or workplace details and account-recovery information.
  • Save screenshots, profile details and message timestamps before blocking the account.
  • Use the dating platform’s in-app report category for impersonation, spam or unwanted contact.
  • Tell contacts who may also have been approached if the same profile appears to be targeting multiple people.
  • If threats, extortion or intimate-image abuse occur, contact local law enforcement or a qualified digital-safety organization.

The takeaway

CupidBot’s proposed model combined automated targeting with undisclosed impersonation in an intimate setting. That is why critics described it as frightening and harassing in effect, even though the available reporting does not make a legal harassment finding. Tinder denied that its founders were employees and said the product had never operated on Tinder. The service’s advertised statistics remain unverified, and its current operating status is not established.

Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.