Apple is joining the race to find the AI ​​code that makes sense

This was an exciting week for the AI ​​community, as Apple joined Google, OpenAI, Anthropic, Meta, and others in the long-running competition to find code that suggests AI to remote users. Like everyone else, Apple made a bet.

Apple Intelligence is represented by a circular shape consisting of seven rings. Or is it a circle with an unbalanced infinity symbol inside it? No, this is the new Siri, Supported Powered by Apple Intelligence. Or is it a new Siri when your phone glows around the edges? Yes.

The problem is that no one knows what AI looks like, or even what it’s supposed to look like. It does everything but seems like nothing. However, it needs to be represented in user interfaces so people know they are interacting with a machine learning model and not just plain old searching, submitting, or whatever.

Although there are different approaches to describing this all-seeing, all-knowing, all-doing intelligence, they have coalesced around the idea that an AI avatar should be non-threatening, abstract, but relatively simple and non-anthropomorphic. (They seem to have rejected my suggestion that these models always speak in unison.)

Early AI icons were sometimes little robots, wizard hats, or magic wands: novelty things. But the implications of the former are inhumanity, rigidity and limitations – robots don’t know things, they are not personal to you, they perform pre-defined automated tasks. Magic wands and the like suggest an irrational, mysterious and inexplicable invention – perhaps good for an image generator or creative sounding board, but not for the kind of realistic, reliable answers these companies want you to believe AI is providing.

See also  The man did not give the New York City building to the Lenape Indians

Corporate logo design is generally a strange combination of strong vision, commercial necessity, and compromise by committee. You can see these effects in the logos shown here.

The strongest insight goes, for better or worse, to the black spot in OpenAI. It’s a cold, featureless pit into which you throw your inquiry, a bit like a Wishing Well or an Echo Cave.

Image credits: OpenAI/Microsoft

Unsurprisingly, the committee’s greatest authority goes to Microsoft, whose Copilot logo is virtually nondescript.

But notice how four of the six (five of seven if you count Apple twice, and why shouldn’t we) use cute candy colors: colors that mean nothing but are cheerful and friendly, and lean toward the feminine (as such things are taken into account). design language) or even childish. Soft gradations to pink, purple and turquoise. Pastel colors, not solid colors; Four smooth, never-ending shapes; Confusion and Google have sharp edges, but the former suggests an endless book while the latter is a happy, harmonious star with welcoming concavities. Some also animate during use, creating the impression of life and response (and attracting the eye, so you can’t ignore it – looking at you, dead).

In general, the intended impression is one of friendliness, openness and undefined potential – as opposed to aspects such as, for example, expertise, competence, decisiveness or creativity.

Think I’m overanalyzing? How many pages do you think the design process documentation was run for each of these logos – more or less 20 pages? My money would be on the former. Companies are obsessed with this stuff. (But it somehow misses the hate symbol dead center, or creates an inexplicable sexual feeling.)

See also  Samsung is building a “digital health ecosystem” with universities and academic hospitals

However, the point is not that design teams in companies do what they do, but that no one has been able to come up with a visual concept that unambiguously says “artificial intelligence” to the user. At best, these colorful shapes convey a negative concept: that this facade is just that no e-mail, no Search engine, no Note application.

Email banners often appear as an envelope because they are (obviously) email, in theory and practice. The more general “send” symbol for letters is pointed, and sometimes split, like a kite, to indicate a moving document. Settings use a gear or wrench, indicating the repair of an engine or machine. These concepts apply across languages ​​and (to some extent) generations.

Not every icon can clearly indicate its corresponding function. How can we refer to the word “download”, for example, when the word varies between cultures? In France, one copy is downloaded, which makes sense but is not really a “download”. However, we come across an arrow pointing downwards, sometimes touching a surface. Down. Same with cloud computing – we have adopted the cloud despite it being a marketing term for “a big data center somewhere”. But what is the alternative, a small button for the data center?

AI is still so new to consumers that they are told to use it instead of “other things,” a very general category that AI product providers hate to define, because doing so implies that there are some things AI can do and some it cannot. They’re not ready to admit it: the entire fantasy depends on AI being able to do anything in theory, it’s just a matter of engineering and computing to make it happen.

See also  Aura Strap 2 review: Context - you'd love to see it

In other words, to paraphrase Steinbeck: all artificial intelligence regards itself as temporarily awkward artificial intelligence. (Or should I say, it is taken into account by the marketing department, since the AI ​​itself, as a pattern generator, does not take into account anything.)

In the meantime, these companies have yet to name it and give it a “face” — although it’s refreshing and exciting that no one has already chosen a face. But even here, it’s subject to the whims of consumers, who dismiss GPT version numbers as an oddity, preferring to say ChatGPT; Who cannot relate to the “Bard” but yields to the “Gemini” who is tested for concentration; Who never wanted to use Bing for the thing (and certainly don’t talk to the thing) but wouldn’t mind having a co-pilot.

For its part, Apple has taken a shotgun approach: asking Siri to query Apple Intelligence (two different logos), which happens inside Private Cloud Compute (no relation to iCloud), or perhaps even redirecting your request to ChatGPT (no logo allowed ), and the best evidence that the AI ​​is listening to what you’re saying is… swirling colors, somewhere or everywhere on the screen.

Until AI itself is better defined, we can expect the symbols and logos that represent it to continue to be vague, non-threatening, and abstract forms. A colorful, ever-changing blob wouldn’t take away from your job, would it?

Leave a Reply

Your email address will not be published. Required fields are marked *