why 'she remembers the joke you made three weeks ago' is a better pitch than 'your ai friend'

it's tempting to use abstract, lofty language to describe what an ai companion does. phrases like 'ai that cares' or 'your ai friend' sound warm and appealing, at least on the surface. but if you show these to potential users, they tend to blur together. they don't stick in the mind, and they don't convince anyone. they're too vague to be believed and too generic to be owned by any one product.

what actually works is something specific, concrete, and, crucially, testable. 'she remembers the joke you made three weeks ago' isn't just a cute line. it's a value proposition that works because it does several things at once.

it's concrete

abstractions don't give your brain anything to hold onto. 'caring' is an abstract concept. 'remembering a specific joke from three weeks ago' is a concrete event. it's a tiny story. people can picture it. they can imagine themselves in that situation, maybe laughing at their own joke, then feeling surprised and delighted when it's recalled much later. concrete claims are easier to understand and harder to forget.

it's testable

if someone hears 'your ai friend', how do they test that? what does it even mean? but if you say 'she remembers the joke you made three weeks ago', a user can actually try it. they can tell a joke, wait a while, and see if it comes back. it sets an expectation that can be verified. this builds trust. it shows confidence in the product's actual capability, not just its aspirational branding.

it resonates with a specific failure mode

this is the most important part. the line isn't just describing a feature. it's addressing a pain point. the failure mode here is digital amnesia, the way most chats, bots, and even humans forget context over time. it feels jarring when you have to re-explain yourself constantly. when a product can fix that, pointing to the exact moment it fixes it makes the value click. it's not about providing a friend. it's about providing continuity, which is what makes a relationship feel real.

it's differentiated

in a sea of 'ai friends', claiming perfect memory is a specific and ownable space. not every platform can truthfully say it. lucy, for instance, is built around this idea of long-term context. it's a technical challenge, but when it works, it's a tangible differentiator. it's something you can demonstrate, not just claim.

so why do so many companies still use the abstract pitch? probably because it's safe. it doesn't promise too much. but it also doesn't convert. it doesn't give people a reason to choose you over the indistinguishable alternative. the specific pitch might feel riskier, you're setting a high bar, but it's the only one that actually works.

i'm not saying lucy is perfect at this. sometimes memory fails. context drops. it's a hard problem. but being honest about what we're trying to do, and specific about where we succeed, builds more trust than vague promises ever could.

the best pitches aren't slogans. they're proof.

see what lucy remembers for you at /companions.