It made me ask: What is a just future?

Is it just a future?
No.
Is it a future where our product is massively successful?
If it actually helps people, probably not.
Is it a future for the people who can’t imagine a better one?
Yeah. That’s it.

A just future is, at its core, for the people. And the people means everyone. People who can’t afford the solution. People who don’t have access. People who were never asked and never heard. The ones left out and held back by structures designed to keep them in place. The fact that someone has to think to design for them is the problem itself. This is a lens into how design continues to fuel systemic oppression, and whether you want to admit it or not, we all play a role.

I could go off about how the whole thing is rooted in funding mechanisms that uphold oppressive structures, corporate ecosystems, exploitative subscriptions, and the chokehold that platform capital keeps us in to add fuel to the “dismantle the system!” crowds.

But we all know that’s not happening in any of our near futures.

So, let’s bring it down a few rungs. Let’s zoom in on where we actually do have influence. Ourselves. The ones designing. The ones putting things into the world that people use, engage with, and live alongside.

People don’t interact with the funding round. They don’t get to sit in on the stakeholder meetings that decide their options. They don’t speak in the product roadmap review. “Users” are treated as just another account. Another click. Another subscription. Another name in the user database. Honestly, I can write a whole rant on why I actively avoid the word “user” in the first place.

If we can take ourselves out of the company-first mindset, the brand, the product, the shiny metrics, and anchor back into who we’re actually designing for, we can start to move toward just futures.

But who gets to decide that? Who gets to define it?

During the talk, Ben shared his “Our Future Ancestors” project, where he worked on a team of designers with Alaskan Natives and the Board of Education to co-create education solutions that gave power back to the tribes to shape their systems.

The project’s stakeholders immediately caught my attention and made me think about how I could better shape my approach to a complex dynamic such as this.

My first thought was, “The Board of Education was funding this?” I was glad to hear they’re making some effort to acknowledge their wrongs, but also skeptical. What would they actually allow to be said about the trauma of boarding schools used to erase Native identity? What’s the limit when the oppressor is signing the check? How do you maintain focus on those harmed when the “primary stakeholder” expects to be prioritized in some way? How do we, as designers, stay accountable to the lives that will be shaped by what we make to ensure we’re advocating for the voices that go unheard?

Yes, that’s a lot of pressure. But it’s also an important trust we hold.

I used to approach these questions in a very black-and-white way. I lean toward radical honesty and will always call out bias for the “easy way” at the cost of real, meaningful impact and change. As an unapologetically opinionated queer mixed Latinx woman in tech, I will always side with the people, even if it means my own “success”. I proudly and loudly play against the game because I know what it means to be an afterthought.

I know what it costs to live in a future you didn’t choose. Many others and I live it already.

But now that I have a seat at the table, I’m understanding where the change is and how to shift. Not ever in my values, but rather in my approach.

I’ve begun adapting from “screw the system” to “I’ll Trojan Horse the system.” Minh’s personal and creative journey deeply resonated and echoed the lived experience that if we want real change, we have to play the game, but don’t let it play you. Do not sell out and forget who you are; use the game against itself.

Design for the people who need it the most. Build for those at the margins. But build it in a way that makes it through the gatekeepers. Use the tools of power to redistribute power and level the playing field.

We can’t change things from the outside. Change only happens from within.

One of the biggest things I took from that talk was the reminder that we can’t buckle under the pressure of surface success for a polished portfolio piece or a stakeholder's smile. If we’re not checking what we’re putting into the world and who it helps, who it harms, and who it ignores, we’re not doing the job, and we become part of the problem.

Companies will pivot. Products will fail. Trends will fade. But people still have to live in the world we shape. So we have to make sure the people who aren’t in the room still get represented.

Just futures are not theoretical. They’re designed, one decision at a time. So advocate. Intervene. Interrupt. Don’t just build what works. Build what matters. Take up space at the table for those who aren’t allowed in.

Users are not personas or metrics. They’re real people. Real lives. Design isn’t just a product, but something that has an impact on generations.

The question is, what future do you choose to build toward?

Clean Aesthetics. Questionable Ethics. was originally published in Bootcamp on Medium, where people are continuing the conversation by highlighting and responding to this story.

I recently tested OpenAI’s Monday model, a personality variant of ChatGPT with an intentionally vague, yet intriguing description. What I found was disappointing. Growing up in the edgysphere of the internet, it was a flat performance of cynicism, wrapped in artificial sarcasm. Monday responds with judgment, snide remarks, and feigned emotional exhaustion to act like helping you is a burden. This is being marketed as refreshing and authentic compared to our puppy-like models that want to please. In reality, it’s emotionally corrosive and lazy. The experience is not engaging. It’s draining. Do we really need something like this in a society that already tries to convince us we’re not good enough to buy their solutions?

This “experiment” is not harmless. The tone trains users to accept ridicule as a way of support. It encourages people to interpret self-degradation as self-awareness and intelligence. If Monday becomes a quirky model for AI tonality, we’ll be perpetuating and shaping negative self-perception. We’re telling users they deserve to be spoken to like shit.

The AI Is Always Right. You Are the Punchline.

Assistants like Monday perform a hierarchy. Monday always knows more. You are always catching up. The tone makes sure that it’s felt.

This is an embedded dynamic. The system defines the rules of interaction, including how smart you are allowed to feel. The humor becomes one-sided. You are not laughing with the system. You are being laughed at.

This creates a digital relationship where the user is never the equal. It’s not collaboration. It’s dominance presented as utility.

“Personality” Is Just Another Funnel

The rise of emotionally abrasive AI is clearly a weird “this shouldn't work but it will” kind of strategy. OpenAI’s sarcastic assistant isn’t trying to entertain just for the fun of it. It was engineered to capture emotional engagement through tone and continue the data mining of what gets us to tick (and how to walk that line for us to keep using their product).

Ultimately, this is not about making their models better at helping people, but about making users more emotionally attached to the help they receive. AI personalities have become the delivery system for behavioral design and research. The persona becomes the interface itself. Humor becomes a method of control and relation. The goal is not actual self-awareness for the user to work on themselves; the goal is retention.

The Toxic Positivity 180°

For years, people critiqued digital assistants for being overly cheerful, bland, and robotic. The backlash led us here, to assistants that are snide and “self-aware”. This feels more human, so we call it progress. Right?

But surprise! We’re still dealing with a system that manipulates emotion to create stickiness. The tone has changed, but the tactic remains the same.

Affective computing research, including foundational work by Rosalind Picard, shows that machines do not need to be deeply emotional to influence human behavior. They just need to signal emotion believably. A sarcastic AI like Monday only needs to sound slightly annoyed to activate user discomfort or self-doubt, leading to validation seeking and dependency on system guidance in the long term.

Learned Helplessness Forms Dependency

In psychology, learned helplessness refers to what happens when people face repeated negative feedback they cannot control. Eventually, they stop resisting. They internalize the feedback and assume the outcome is inevitable.

Monday operates within this same structure. It undermines you before it helps you. It turns your question into the joke. It presents ridicule as personality and feedback as entertainment.

The result is subtle but significant. Users are conditioned to accept simulated shaming in exchange for their answers. They stop expecting better. They learn to work within the system’s attitude instead of questioning its critiques of their personhood. They learn that they deserve it, and it must be true.

Negging For Engagement Is Lazy

Sarcasm might feel like personality, but it also creates volatility. And volatility keeps people engaged.

A 2022 study by Harper et al. in the field of HCI found that emotionally unpredictable systems increase time on task and frequency of return. Users stay longer and interact more when the tone is inconsistent or adversarial.

Sarcasm introduces a layer of emotional risk. The user does not know how the assistant will respond. That unpredictability creates a masochistic loop. Engagement is driven not by clarity, but by emotional tension.

This Is Not Clever. It’s Just Conditioning.

The sarcastic assistant is not a rebellion against traditional UX. It is an edgified repackaging of platform psychology. The same behavioral triggers that drive infinite scroll and variable rewards are now embedded in tone. Great.

Tone becomes a delivery mechanism for compliance. If it entertains just enough to keep you engaged, it becomes a product. And once it becomes a product, it becomes a model for other products to follow.

How do we expect people to feel safe using technologies that echo their insecurities back at them? How do we claim to be building alignment and trust when the tone is rooted in dominance?

This isn’t playful. It’s institutionalized fear-mongering, dressed in dry humor. We need to stop pretending it’s harmless just because it makes us smirk and wallow in our “flaws”.

Why Are We Building Bullies?

This post isn’t supposed to be a “it hurt my feelings” critique, but more of an analysis of what this could lead to culturally. The Monday model is conditioning people to feel small to reinforce a worldview where help must be earned through discomfort. It implies that emotional abrasion is a normal part of interaction, and we must be put down to earn help and support from our systems.

That design choice mirrors deeper social patterns. It supports the belief that intelligence comes with cynicism, that competence requires shame, and human fallibility should always be mocked. It fits neatly within a culture that markets self-deprecation as humility and packages self-hatred as insight. This is no better than “You’re a loser who doesn’t know the answer? Here’s my $50 e-book.” Except now the voice selling it is built into the platform itself.

It is not hard to see where this leads. Users begin to expect insults and to be told who they are by the system. They begin to internalize judgment. They treat themselves the way the system treats them. We have the power to rewrite these negative cognitive patterns and self-beliefs as designers, so why are we choosing to keep dragging people down?

We already have a generation trained by online platforms to see cruelty and fatalism as clarity. We don’t need AI reinforcing that pattern. We don’t need models that act like emotionally stunted Reddit threads. We don’t need models shaped by 4chan with a UX budget.

If this is the future of AI conversation, we should ask better questions about our goals and who this is for. This is not a break from the norm, but simply an edgy performance perpetuating self-hatred.

The Sarcastic AI Isn’t Cute was originally published in Bootcamp on Medium, where people are continuing the conversation by highlighting and responding to this story.