And the winner of "word of the year" of the year for 2022, as chosen by writers in tech (and not by robots) is … gaslighting!
Permacrisis came in a close 2nd, with goblin mode not too far behind. Homer struck out and there was even a write-in vote for "Huh? What? Was this really a thing?", which likely captured how many felt.
OK, enough WOTY already—bring on the AI! This special issue is a deepdive into AI, so grab a snack and get comfy.
SPECIAL AI EDITION ✍
Issue #214 (Dec 14th, 2022)
AI: the good, the bad, & the terrifying 🤖
AI meets UX 🦾
AI: where are we now? 🤷
Articles, microcopy, and conversations 💬
For the past few years, we’ve been following the development of AI, particularly AI writing tools and how they might affect our field. Are they a fad, a gimmick, a fancy toy we’ll soon grow tired of? That’s almost certainly not the case.
But how will they impact professional writers? And is AI itself a threat to human civilization?
We read through tons of AI articles so you don’t have to. In this issue, we’ll let the headline and bullet points tell the story, and give you our take on the matter.
(Created using Midjourney)
WHY NOW? ⏰
Even if you have been living in a cave, you’ve probably heard of OpenAI’s new chatbot, ChatGPT. Using the new generative text mode davinci-003, ChatGPT can understand complex instructions and produce remarkably high-quality long-form texts, rhyming poetry, and even functional software code.
It is game-changing technology that, in an instant, became available to all, and people from all fields are now scrambling to process the implications.
The robo-cat is out of the bag, and there’s no putting it back in. AI is now part of our lives in a direct way.
AI: THE GOOD, THE BAD, & THE TERRIFYING 🤖
Artificial intelligence dates back to the 1950s but has recently seen exponential growth in its abilities. And while it’s been a boon to fields like science and medicine, the potential for misuse looms large with consequences ranging from troublesome to downright terrifying.
Google’s deep-learning program for determining the 3D shapes of proteins stands to transform biology.
The ability to accurately predict protein structures from their amino-acid sequence would vastly accelerate efforts to understand the building blocks of cells and enable quicker and more advanced drug discovery.
Using new machine learning techniques, researchers have developed a virtual molecular library of thousands of "command sentences" for cells, based on combinations of "words" that guided engineered immune cells to seek out and tirelessly kill cancer cells.
The potential benefits of AI are limitless. Yet many warn of its dangers including deepfakes and synthetic media that could increase disinformation, erode trust in institutions, and hinder our ability to distinguish between what’s real and what isn’t.
And, not to alarm you, but some predict AI will cause massive economic disruptions and even threaten human civilization. So there’s that.
Some critics fear that [OpenAI’s tech] could be our undoing, especially with more sophisticated tech reportedly coming soon.
"Shame on OpenAI for launching this pocket nuclear bomb without restrictions into an unprepared society … ChatGPT (and its ilk) should be withdrawn immediately. And, if ever re-introduced, only with tight restrictions."
"The most disruptive change the U.S. economy has seen in 100 years," and not in a good way.
People are going to be forced to make disclosures that ‘We wrote none of this. This is all machine generated.’
The purpose of writing an essay is to prove that you can think, so this short circuits the process and defeats the purpose.
In 2021, cybercriminals used AI voice cloning to impersonate the CEO of a large company and tricked the organization’s bank manager into transferring $35 million to another account to complete an "acquisition."
Threat actors can not only mimic an individual’s physical attributes to fool human users via social engineering, they can also flout biometric authentication solutions.
The app, which prompts users to upload photos of themselves then spits out 50+ AI-generated portraits, has been decried by artists as predatory to real, human-made artwork.
Lensa AI is able to transform innocent photos of fully-clothed women or children into AI-generated nudes.
Generative art tools can be lots of fun—we’ve been using them at the HUB for a while now. And the rate at which they’ve improved is crazy! In a few short months, we went from this avocado chair (Jan ‘21):
To this (Dec ‘22):
Yet, it still has limitations. For instance, it’s very difficult to reproduce the look of a custom character or create a series of consistent images. For one-off images, it certainly threatens artists. But when it comes to things like illustrating a book with consistent characters and style, the technology falls flat.
GPT-3 is a computer program that uses advanced algorithms and a vast amount of data to understand and generate human language. At least, that’s what ChatGPT said when asked to explain GPT-3 at a 7th-grade reading level.
There are now many AI writing tools on the market, and we at the UX Writing Hub have used them in the past. But does this new davinci-003 model represent a sea change in how writers will work?
Letting an AI write product reviews (for an iPhone 14).
The flow of the review is textbook; ChatGPT sets the tone with what it thinks about the iPhone 14 Pro, talks about the design, performance, cameras, and battery life, and even concludes with a definitive recommendation.
The main problem with the AI-generated product review is that it's inaccurate.
Review Metallica's album Master Of Puppets favorably, emphasizing the bass playing of Cliff Burton.
Review Metallica's album Master Of Puppets unfavorably, emphasizing the drumming of Lars Ulrich.
Both reviews were quite good.
AI: WHERE ARE WE NOW? 🤷
The technology is impressive—staggeringly so. AI can also write software, film scripts (complete with character descriptions and dialogue), and is getting better at complex games.
There’s no doubt that AI writing tools can help us improve our workflow. Yet, AI-generated texts still have a big problem—they are, far too often, full of shit. They lie, they B.S., they make stuff up. And what’s more, they do so without knowing that they’re lying.
This is because, as impressive as the technology is, AI cannot yet truly think, at least, not in the way humans can.
And since writing is, above all, thinking, AI cannot yet replace human writers—it is not currently a threat to most writing jobs. And whether or not AI will ever be able to think like humans do is still up for debate.
In the meantime, one thing AI can do is write blog posts that are, in some cases, good enough for SEO purposes. So SEO-focused content writing is probably the most threatened writing job for now.
As for misinformation, B.S. on the internet is nothing new—people have been dealing with mis- and dis-information for as long as we’ve had written language. And although AI-generated texts will likely bring an increase in misinformation, perhaps they will force us all to become better judges of information, vigilant and hyper-aware of the need to carefully discern fact from fiction, truth from lies, news from fake news, human from robot.
Perhaps, ChatGPT will give our collective B.S. detector a much-needed fine-tuning.
FEATURED ARTICLE 📰
Though most of our UX jobs are safe for now, we must start preparing for the future. UX writer/content designer Sophie Strosberg shares what she’s learned about putting a human touch on your work.
Here are five ways to show your humanity in your work and stay ahead of the AI curve.