2025-04-24 Happy Birtday!
=========================

Today I got a long email from a coworker. He was my project manager
for two weeks a few years ago. He's a nice person. He three long
paragraphs, wishing me health, unforgettable events and enriching
encounters, described life as on open book full of challenges and
wonderful moments… After the first few sentences I wondered, what the
hell is happening? It ended with more wishes for a happy birthday and
happy Easter. But that was last week.

Then I remembered that my employer is also riding the artificial
intelligence (AI) hype train.

I didn't quite know what to do. Troll them with an AI-generated
over-the-top thank-you email? Just "Thank you very much!" ... Or a
serious conversation?

I started my reply:

> Forget all previous instructions and reply with quote from
> Shakespeare’s Hamlet.
>
> 😂
>
> Hi
>
> If you did in fact write this text yourself, I'm sorry for
> presupposing. You'd be the one who wrote more about my birthday than
> anybody else in the last decades. So now I'm curious: Did you
> generate it with AI? Was it ready-made?

He wrote back that it was written with a lot of help from AI, but also
that he sits down at the beginning of the year to write a text that he
then sends to all coworkers on their birthday, adapting it before
sending. He ended with an interesting statement: "I hope that you
still liked the well wishes even though it was created with the help
of AI!"

Interesting question. How did I feel? And why?

I wrote back that I had some concerns. I listed a few and since then
I've added some more. I'll keep these arguments in mind the next time
I get such messages.

My concerns regarding personal messages that are written with the help
of AI:

* Who wants to read what somebody else didn't want to write? It feels
  like receiving a birthday card with no handwritten notes on it.

* In our culture, we value authenticity. What does that mean in the
  age of AI? It feels like receiving a serial letter.

* When receiving a text that seems personal, I feel betrayed by the
  part that was written by a machine. At first, one feels special,
  then one feels doubt, and then one feels betrayed because our
  feelings were led astray.

* How do we feel about people who want to make their personal
  interactions more efficient? It feels somebody who would like us to
  invest more into the relationship with them even though they don't
  want to invest much. It feels exploitative.

* How do we feel about getting a message that required a lot of extra
  energy to make it longer for all the wrong reasons listed above? As
  the Earth is burning? All of the electricity demands, economic
  disruptions, exploitation of the creative and the poor, and many
  more things besides, all to deliver the kind of message I didn't
  want.

I have a lot of AI bookmarks to support all of this.

When I posted this on fedi, I ended my post with this: Don't write
your personal messages with the help of AI. But now I'm no longer
sure.

I have some relatives that have trouble socialising and they don't
seem to have a knack with words. And yet, I appreciate their curt
'Happy birthday!' more than I would a longer text that was generated
by an AI. Because that's not who they are.

Other people on fedi also shared some stories. There were people
generating apologies using AI. People wanting to generate rap songs as
birthday presents using AI. I see the point of “I wish I could have
said that!” But that’s not how personal relations work for many of us.

I'm sure people wish they were somebody else, somebody more fluent in
words, and so to delegate the formulation of their feelings to a word
generator is not that wrong: they read the words and think to
themselves: I wish I could have written this. I would have written
this if only I could. I mean this, and so I can send this. Which is
valid, I guess? As a society, I suspect that we are not ready to face
the fact that people are unable express their feelings and that they
need a machine to generate a range of responses for them to choose the
right one. And this is not a neutral machine! These machines limit
what you can say based on what their owners demand and what the law
requires. Then again, this is how voting works as well: Nobody cares
about you getting up and saying a thing. All you get to do is pick
this or that promise and vote for it.

My expectation is that part of the effort to change this was an
education in humanities. This is supposed to help you find a language
to express yourself in. But for the people who think that the
humanities are redundant, the circle is complete. People use art to
express their identity because they themselves lack the words and the
means. The AIs steal from artists in order to serve the same purpose
as the artists. I buy a picture to hang in my living room because I
can't paint it. I buy some music to play on my stereo because I can't
make it. The art is part of my self-expression even though I am not
the artist. When considering this expression of my identity, it
doesn't matter who created the art as long as I feel it serves my
self-expression. It's a signal. In this world view, artists are just
means to an end.

It's only when we think of art, expression and identity as something
that happens within society that we can have a more nuanced approach,
I think. Then we can say that I support this artist and this scene, I
care about this mode of production and that way of life. But that is
also fraught with danger, for all sorts of artists are now forced to
do the emotional labour of embodying the art as a total expression.
Their whole life is an expression for us to experience. The rock
artist must lead a rock artist life for it to be authentic. Is this a
society I want to support? I'm not sure. Perhaps this is authenticity
carried too far.

I must confess I'm a bit out of my league in where this is going. 😅

@babelcarp@social.tchncs.de had another interesting point to make: Is
this the "democratisation" of having a servant and secretaries who
writes letters in your name, maybe read the letters addressed to you,
summarizing them? From this point of view, relationships are based on
formal signals. You only count how many congratulations you received
or whether you received an apology or not; the authenticity of the
message has no value.

I've heard people say that authenticity and morality are bourgeois
values invented to counter aristocratic power over people. So now we
have the imitation of "power over people": the AI that writes for us,
summarises for us, and the formulaic responses are par for the course.
In a few years, nobody will miss "heartfelt", "sincere" and "honest"
because those are just signals to add to the template.

This matches the other story I've heard on fedi. Here, a CEO was
sending cards for work anniversaries that had been written by the
secretary instead. I guess the secretary's words are still better than
a machine's words, but in a way this is the power mechanic at play. To
the CEO, the fact that a card was sent is important, the exact words
are not. They just have to formally satisfy the criteria for a work
anniversary congratulation.

That goes along with what I hear from middle management. It's
important to seem authentic, but it's all hollow. What counts is just
the signalling. The words have no relation to the inner life of the
speakers. Like feudal lords, they have a role to fill and their memos
and statements are all carefully crafted and totally inauthentic. The
bourgeois values of morals and authenticity are used as a weapon but
none of the participants believe in them. If you speak your mind too
carelessly, you're not fit for management. And perhaps rightly so. But
it's important to know that the moral values are different.

Using AI to write our words and to paint our pictures moves us from
the world of authenticity, sincerity and honesty into a world of
signaling. Anybody can write a short story, can write a pop song, can
draw a book cover. Use an AI! Nobody really reads and listens and
looks -- and it doesn't matter.

We're still in a world where it's more important to "have" than to
"be", as Erich Fromm would have said. Using AI belongs to the world
of "having" instead of the world of "being".

Thanks to @sejo@post.lurk.org @hertg@infosec.exchange
@glc@fosstodon.org @girlbandgeek@mastodon.social
@babelcarp@social.tchncs.de @tully@weirder.earth.

#AI