February 13, 2024

You Vs. The Guy She Told You Not to Worry About

You Vs. The Guy She Told You Not to Worry About

Or A rambly, self-indulgent and indignant retrospective on Dear Ai

Introduction

Looking for a little help with your dating game? RIZZ has you covered! Our AI-powered app uses cutting edge LLMs to generate personalized responses that are sure to impress your crush.

RIZZ - Launched 2023

Last year I devoted myself to a Design Fiction project called 'Dear Ai'. Dear Ai was a 'speculative' company that allowed users to generate personal letters and send them as faux handwritten cards. It told of a near 'future' where Generative AI would further erode what it means to be human. Like many of my Design Fiction work, I chose to place it in front of the public as if real, with little context except for an about page¹. I truly believed, and still do, that a 'future' where people send AI-generated sympathy letters to their friends, that they customise by deciding how sad they are for their friend on a scale of 'not at all' to 'extremely', is fundamentally unsettling and a 'future' we should move away from. I felt certain, I do not any more, that people shown this 'future' would be provoked and contemplate their relationship to technology or think about how we assign old notions of value in this changing world and decide for themselves what machines should do, in a time when we are told they can do everything.

So imagine my surprise when I received an email from an Animal Hospice administrator telling me Dear Ai's letter generator was not working. They seemed slightly frustrated at me because they wanted to try the free plan before committing to the pro plan in order automate their doggy death letters, I assume.

Launch Day

We're here to diversify the fashion industry and challenge the status quo when it comes to inclusivity, sustainability and innovation.

Our software platform seamlessly integrates with Browzwear VStitcher, enabling you to showcase your 3D designs onto our industry-leading generative AI-models. Watch your creative vision come to life, with a range of customization options at your fingertips.

LALALAND - Launched 2019

The first hint of Dear Ai's impotence became apparent early on after launching. Like all of my Design Fiction, I launched it on LinkedIn. A space where I feel my uncanny strand of near-future Speculative Design can catch people off guard and affect them in a way work placed in museums, with its distance from real/lived life, can't. This seemed like a good plan. After all, I used the same tactic for the launch of my fictional Venture Capital company Ventually and that had gone pretty ok. Ventaully seemed like a good model to follow being that it was also a Nick Foster 'Future Mundane' inspired project and I always saw Dear Ai as just another company that Ventually might fund. What's more, I had tried to apply a lot of the lessons I had learned from Ventually to Dear Ai². However, whereas Ventaully's launch post garnered a moderate, if at the time underwhelming, amount of engagement, a lot of concerned friends and confused family, Dear Ai was met with a resounding 'meh'.

Equally, when I ran Facebook adverts for Dear Ai, another tactic I use for placing critical objects in everyday life, my £50 advertising budget got me less engagement than a bad busker on the underground. Again compared to the flood (a small £50-sized flood) of angry comments and shares that the Ventually adverts incited, I was confused, and a tad bummed out. I'm not saying that the worth of a piece of Design Fiction is how many people spit out their morning coffee in indignation or leave disgruntled comments, but I do believe it's a good indicator that your work has provoked and that hopefully, people are engaging with the ideas that informed the work. After all, to me, the mark of good Critical Design is its ability to spark some kind of debate or discussion around a topic through the design of some insightful artefact. In this respect, it's fair to say Dear Ai was not doing great. With hindsight, it seems so obvious now that this was always going to be the case. After all, there were so many signs.

The Signs

Personal AI is a digital extension of anyone’s personal brand.

It is an AI model that uses your own data, facts, and opinions to create a responsive and interactive messaging experience that helps people be more productive and deepen relationships. Think of it as your very own “Hey, Siri” that can respond using your own facts, opinions, and style.

Personal.ai - Launched 2020

To start with the project was conceived as a follow-up to another Design Fiction project I did a few months earlier, called Dear-MP.UK. Dear MP offered UK residents the ability to send complaint letters to their MP in seconds. All they needed was a UK postcode and something to complain about, Dear MP would take care of the rest. I had hoped it would start discussions about the use of generative AI in propaganda and censorship through noise, maybe highlight the changing nature of value in a post Open AI world or help articulate how Generative AI is a technology that normalises and further entrenches outdated models and standards. However, the project was met with a warm reception, with many Service Design friends telling me how tech like this (Dear MP) would democratise this exclusionary style of formal letter writing. I, not expectedly, disagreed with this being a good thing³ and felt frustrated that my project didn't have the impact I desired. Others further added to my disillusionment by suggesting that this use case of AI was in the sweet spot for AI being good enough at being good enough. I thought perhaps people didn't see my critiques because the topic of government is distant. Who cares if an MP gets 1000 letters a day generated by AI, it's part of their job and makes my life easier, kinda of thing. Maybe people would be more affected if I showed how this same technology could disrupt their personal lives. What if rather than the MP receiving an AI letter about potholes, it was you receiving an AI letter saying Happy Birthday from your Brother? So I started working on Dear Ai and tried to forget the definition of insanity.

The second sign was all the kind advice I got from my friends when showing them Dear Ai before the launch. They told me many a time, in lovely constructive ways, that there wasn't much scary, provocative or impactful about my project. I would reply to them, 'What do you mean not scary? It's a company offering to automate all your personal communication with AI?!? The example text for a sympathy letter is, "Their child passed away!" The website says it will integrate your social media data to make your letters more personal! What do u mean it isn't provocative enough?'

Stubborn as I am, these kinds of conversations did impact the final output. They are what convinced me to add Love Letters as a letter generation option, alongside the Birthday Cards, Thank You Letters and Sympathy Cards. I had originally put this off because I felt that Love Letters was too on the nose, that it would take my Design Fiction from believable and, in my theory of Design Fiction, therefore impactful to implausible and therefore dismissible. However, even after this addition, still more friends told me it was 'fun' and 'useful'. The former adjective being not exactly what you want to hear when you're trying to be serious and the latter sending shivers down my spine as I recalled Dear MP. Blinkered as I was then, and kinda still am, I decided these smart Critical Design adjacent friends aren't my audience, they already know too much about AI, they are already so cynical, blah blah, what does it matter what they think...

So I launched it and as mentioned before, it barely disturbs the water. It's not as if no one finds it, traffic wise the project has been a success. Real people actually use it, and they are not only generating letters but are in fact, sending them to other real people. The main criticism seems to be that the letters Dear Ai generates are not good enough, or that there should be more languages available. So I have done it again. I have made something that failed to cause discussion but is really seen at best, as boring, and at worst, as useful. I have done the same thing again and expected different results.

Doppelgänger

Create personalized Bedtime Stories for your children.

With the latest AI technology, the personalized stories can now also be transformed into audiobooks. Turn your children's magical adventures into unique listening experiences with pleasant, soothing voices that stimulate the imagination and make falling asleep a pleasure.

Oscar - Launched 2023

I remember launch day and how I posted it on Product Hunt, a techy site for people to announce their new startups. That most upvoted thing on the site that day was a tool that used AI to generate user personas, a thing usually created through interviewing people and hearing their unique voices. I thought to myself is this also a Critical Design project; perhaps riffing on what it means to be human when we keep personifying AI? No, it turns out it's a real app and people use it. And that's when I think it hit me, my Design Fiction was indistinguishable from the current reality. OK, maybe that's parabolic, but it's a lot damn closer than I feel comfortable with.

Some may argue that that is not a surprise since the third warning sign of my fart in the wind of a project was that I sort of plagiarised the idea from Microsoft's Bing AI launch. Where during the keynote they demoed using Bing Ai to email your family⁴. However, here's the cherry on top of it all. Yesterday, I found a real, Bloomberg news covered, company that is using AI and robots to automate handwritten letters. In fact, I found 2⁵. My dystopia is not even some Venture Capitalist's Utopia, it is all of our reality.

-

I wonder if the first symptom of getting old and out of touch is scrolling through X (formerly known as Threads) and not knowing if a blue checkmark's tweet is a parody or not. This very morning I read a tweet that said 'As a consumer would you rather have one personal agent that can book your travel, order you food, help manage your finances, manage your shopping, find you entertainment, keep in touch with friends, track and manage your health, and teach you things OR one specialised for each OR a useful friendly agent on every site you visit?'. I spat out my morning coffee indigently, shouting at my phone that one of those things is not like the others. You can't just sneak in 'Keep in touch with friends' in a list of things AI should do. However, looking at the replies everyone just said they wanted option 1, so maybe you can.

We live in a time when tech company launch announcement feels to me like Charlie Broker's discarded Black Mirror scripts. Where Apple advertises the Vision Pro's Spatial Video feature with a clip that shows two young kids playing on the floor and their dad ogling at them through an insect-like ski mask as he takes a Spatial (3D) Video of them. His eyes hidden and replicated by technology, face covered and seeing his children with a slight digital delay⁶. Where Humane's Ai Pin launch video has the world's least charismatic presenter asking his AI Pin to make his text message to his friend 'sound more excit[ing]'. Or where the newest Google Magic Editor feature allows anyone to Eternal Sunshine themselves by editing their photos into a different reality by changing the weather, people's faces, and background. I could go on, but I think my point is made. My criticisms are their selling points and my art their flagship product.

So why am I surprised when my seemingly dystopian project appears on Superhuman's list of 5 AI Tools to Supercharge Your Productivity, next to a LinkedIn SEO Tool and an AI video transcriber. Am I out of touch? Most likely, I still swear by wired ear phones. Is the world becoming more dystopian? On my slice of Twitter, the consensus is no — in my local pub, very much yes. Do I know how to move forward? Not entirely but I have a few options.

The Options

Handwrytten’s technology allows you to scale your handwritten outreach, creating positive impressions and long lasting bonds.

Handwrytten - Launched 2021

To preface this, I am not saying Design Fiction is a dead medium. I am not so proud that I believe my failure is a reflection of the whole discipline. I am highly aware that it might just be that I made a couple of out of touch, unimaginative and dull projects. In some respects I obviously did. However, I do think that these projects have made me question my relationship with the discipline, my views on Technology in general, and why I do what I do.

In his essay 'E Unibus Pluram', David Foster Wallace worries about 1990s irony heavy literary fiction's inability to be an antidote or a worthwhile critique to TV and corporate culture because they have co-opted irony and rebellion for themselves. 'How can even the idea of rebellion against corporate culture stay meaningful when Chrysler Inc. advertises trucks by invoking “The Dodge Rebellion”?'

It feels the same could be said about Speculative Design, of which Design Fiction is a part. Excluding my issues with not being able to tell the difference between the hot new start-up and speculative art project, the field seems so deeply entangled with Big Tech funding and its professional sibling foresight⁷, that it no longer feels like the same exciting, rebellious and provocative discipline I fell in love with at Art School. Wallace's solution for when art becomes indistinguishable from what it critiques seems to be that we should change our methods. For fiction writers, he suggests one approach of creating work that is intricate, sincere and requiring a lot of its readers in response to TV's passivity and witty people-pleasing nature. Should I do the same? Should I finally shelve my theory that Design Fiction work should be accessible, upfront, and playfully engaging? Should I instead start making abstract work that people can silently ponder over in museums? Maybe, I do want to do some more exhibitions, even If I'm not sure I believe in them as an effective space for Critical Design.

Or I could take Near Future Lab's Julian Bleecker's advice and 'Imagine Harder'. Stop thinking near future or future mundane and start thinking of more elaborate futures. Perhaps even futures I want, rather than ones I find scary. I could even go one step further and take my Mum's advice. Start making real things I want to see in the world rather than writing 3000 word essays on why I think everything is broken. Could do. Although my idea of a better future feels banal and I see glimpses of it already in my everyday life. I see it when I go to my local coop supermarket and check in with the volunteer who is scanning my groceries because we worked together last week. I see it when my local bartender ignores people queuing at the bar because she wants to finish a conversation with an old friend. I see it when my neighbour asks to borrow eggs in the building group chat. As much as I like Raymond Carver, these don't feel like very fun or interesting things to make speculative work about. Nor do they seem to be the kind of things that need a designer to come in and create a product around. God knows we have enough Design-Solutionsism in the world already.

So what am I left with, more complaining? More sentences phrased like questions? No, I think it's just time for me to get off the high horse of Critical Design, and maybe even the word 'Design' and the authority it clings to. I think I need to stop putting so much weight on the (my) intended impact of my work and accept all its interpretations. I want to quit thinking of my work as purely political when it is so obviously also personal, these are my fears after all. My work is my version of Art or more accurately as Zadie Smith puts it, it's 'Something to Do'. It's something I am privileged to be able to divide my time with. I enjoy experimenting, making, being critical, being playful and thinking about the 'future' (even if it happens to be the present). I'm enjoying writing whatever this is. Rather than get down about the state of Speculative Design and feeling trapped by a discipline, I should take everyone's advice and no one's. I should just do more things that I enjoy or think are important, both in my work and personal life. Who cares if people don't care or misinterpret my outputs, or if my Twitter feed feels like the degree show website of a Critical Design course? These are just things to learn from or to make work about. And If any of that work moves the dial in some capacity, which I still hope it does, that would be amazing. In short, and apologies for the lack of character development, I kind of feel I should just keep doing what I'm doing: making weird little websites and writing rambling essays full of spelling errors.

---

Thanks for reading. Sorry it got so long, wasn't my intention. Hope you have a great week,

Fred