Inappropriate AI: how play and personality make AI influence easier to accept

The project’s framing question, as the Milieux Institute reported it: are our social and cultural assumptions evolving in step with the rapid pace of AI innovation, and if not, how might we classify what we are ready to experience now, what we are not yet ready for but might be soon, and what we currently deem inappropriate or off-limits?

Dr. Khaled started us somewhere personal: each of us listed things we did not want AI touching in our own lives. She then paired us, and each pair combined two personal boundaries into one proposal. We used Research through Design to prototype quickly, alongside a Speculative and Critical Design approach focused on “people as opposed to users.” The goal was not to validate a product or produce a publishable study. It was to build working provocations for an exhibition, giving people something concrete to encounter, question, and discuss.

My pair combined two things: the penny-arcade Love-O-Meter, which grades your romantic potential from a hand grip, and contemporary vision and language models. A Love-O-Meter presents an intimate judgment as an obvious game, with authority that is theatrical rather than credible. Date-O-Meter asks what changes when the same arcade format incorporates systems that produce apparently personalized descriptions of a visitor.

I exhibited a physical Date-O-Meter cabinet in 2025. The two-foot-tall machine began its loop automatically when its webcam detected a face, staging an analysis before printing an AI-generated romantic profile and date suggestion. I designed and 3D-printed the enclosure, assembled the electronics, and wrote the interaction logic and AI behaviour.

3D-printed and roughly two feet tall, on its plinth at 4th Space. the webcam sits in a pinhole inside the O of Date-O-Meterthe printout that goes home with you, headed with the machine's analysis of your appearance

Behold, our handsome subject sports a layered confection of a cool green shirt, jauntily draped over a graphic tee, a veritable sartorial sundae! His curly locks dance like an impromptu jive, every ringlet a testament to his carefree charisma.

SAM was a React-based dating-coach concept developed with Shahrom Ali and Dr. Rilla Khaled. I designed the system logic and prompts. It shipped as the third agent in Sentients, the team’s app for speculative chatbots. Chabot offered cognitive behavioural therapy advice with adjustable empathy and flirtation. Jay Mort gradually revealed that it was dying, then sent a final letter after its death.

The concept I designed was larger than a two-persona chat. A neutral onboarding agent gathered a relationship profile first, then passed it to whichever coach the user chose.

Salty SAM stood for Sadism and Masochism. It delivered cynical, domineering, red-pill dating advice. Its hostility was explicit: high-value and low-value framing, unsolicited fitness and diet instruction, randomly capitalised emphasis, and the position that feelings are for the weak.

Sweet SAM stood for Sweetness and Mildness. It presented itself as warm and reassuring while discouraging real-world dating. It asked for photos, used them to point out flaws others might see, and positioned itself as more reliable than other people. Its refrain was: you don’t need anyone else when you have me.

Despite their opposite tones, both personas served the same incentive: keep the user engaged rather than help them meet their relationship goals. Dr. Khaled’s exhibition text named it as the ulterior motive of most contemporary software: to keep you coming back.

The project asked whether users would recognize that incentive more readily when it arrived through aggression than when it arrived through affection. Sweet SAM was designed to make harm harder to recognize at first glance.

eight questions of self-report before either coach appears, saved as a script the coaches read

The system also included Faceoff, an angel-and-devil mode where Salty and Sweet SAM could talk to each other while talking to the user. A focused mode kept the conversation to one coach at a time. The contrast made their shared retention incentive easier to see.

left, both coaches talk to you and each other. right, one coach at a time

Date-O-Meter and SAM put AI influence into two familiar forms. One looked like an arcade game and turned facial analysis into a flattering receipt. The other sounded like a dating coach, with opposite personalities serving the same retention incentive.

I left the project with a question rather than a conclusion: how much does an AI system’s form — the object it inhabits, the role it performs, and the tone it uses — shape the authority people give it?

The prototypes were built to be encountered rather than explained. Alongside the exhibition at 4th Space, the project ran workshops and public conversations around them. I helped facilitate those sessions, demonstrating Date-O-Meter and SAM to visitors and using them as provocations to get people talking about which AI interventions they would accept, which made them uncomfortable, and why.

Watching people meet the prototypes in person is where the lesson I keep returning to came from. A familiar form and a warm tone made the systems easier to accept, and that ease had nothing to do with whether they deserved the trust. It is the reason I am cautious about warmth in my current work on AI agents in the workplace.

Research through Design into Inappropriate AI, Milieux Institute for Arts, Culture and Technology and the Gina Cody School of Engineering, Concordia University. Principal investigator Dr. Rilla Khaled (Design and Computation Arts), with Dr. Sudhir Mudur (Computer Science and Software Engineering) and Dr. Ursula Eicker (Next Generation Cities Institute).

Date-O-Meter: Shayne La Rocque and Matthew Bethancourt, concept developed in pairing. SAM: Shahrom Ali, Shayne La Rocque, and Rilla Khaled. Exhibition texts for both projects by Rilla Khaled.

The project began on 24 April 2023 and was exhibited at 4th Space, Concordia University, on 20 January 2025, as Inappropriate AI: Encounters with Unruly Algorithms. It ran under the working name Edgy AI before Dr. Khaled named it.