Grimes, Grok, and slaughterbots: the highs and lows of AI in 2023

Dictionaries came together in rare agreement over their words of the year for 2023. Authentic. Prompt. Hallucinate (times two). OK, so Oxford University Press threw in a rogue Rizz, but otherwise this year’s contenders honed in on an overwhelming concern about a particular emerging technology and its destabilising effects on our reality. Collins got straight to the point and simply chose: AI.

It’s no surprise that AI is what we’ve all been talking about this year. In the last 12 months alone, the technology has grabbed headlines with applications ranging from virtual girlfriends, to realistic art and music generators, to tiny virtual societies. Unless you’ve been living in a cave out in the wilderness (in which case, welcome back, and sorry in advance) you’ll know that frontier models like OpenAI’s GPT-4 have even sparked concerns about an imminent machine-led apocalypse. Exciting!

Given the pace of AI’s development in 2023, and its unexpected knock-on effects, there’s been a lot to take in. Spooky robots. Corporate melodrama. Waves upon waves of uncanny images. That also means there’s been a lot to miss, but don’t worry. Below, we’ve gathered some of the highlights (and lowlights) of AI in 2023, to get you up to speed.

AI HELPED UNLOCK NEW MEDICAL BREAKTHROUGHS

In the never-ending debate about the pros and cons of AI, the medical benefits are one of the more straightforward wins. Admittedly, you might not be ready to hand your entire life over to an algorithm just yet (like aspiring immortal Bryan Johnson) but it’s hard to deny that it’s a good thing to speed up the discovery of life-saving drugs and match them to patients without arduous IRL trials, or spot diseases like cancer months in advance. 

Using AI to dream up cures for our mortal bodies “could sound like science fiction” said researcher Vanessa Smer-Barreto in an interview with Dazed earlier in the year, but the results speak for themselves. Her research alone – which used AI to develop drugs that can target “zombie” cells linked to ageing – saw a several hundred-fold reduction in drug screening costs, and might have saved years of research.

AI MUSIC HAD ITS UPS...

Leave it to Grimes to spearhead the transition of musicians from mere mortal beings to transcendent human-AI hybrids. In May this year, the musician made herself “open source and self-replicating” via Elf.Tech, allowing anyone to use her voice and likeness in their music. The results came thick and fast, with Grimes herself declaring one track a “masterpiece”.

It’s not just Grimes, though. In August, Sevdaliza also open-sourced her voice via AI, introducing the world to her robot counterpart, Dahlia, in the process (read Dazed’s interview with Sevdaliza here). Elsewhere, patten pioneered the first LP made entirely from AI-generated sound sources, Mirage FM, ushering in a new age of AiDM.

... AND ITS DOWNS

Of course, AI music has its haters, too, largely depending on how the technology is used. Unlocking a new palette of sounds for your own original albums? Cool! Hauling dead musicians from their graves to perform for the rest of eternity as subservient digital avatars? Not so cool! Flooding streaming algorithms with bland, AI-generated muzak that eats into real artists’ profits? Definitely not.

ELON MUSK’S GROK PLUMBED NEW DEPTHS OF CRINGE

When Elon Musk announced that the world needed a “based” AI to compete with “woke” ChatGPT back in March, we only had one thing to say: Concerning. What we didn’t know back then is how the billionaire’s chatbot Grok, unveiled just a few months later, would achieve levels of cringe previously thought impossible. See: a sense of humour lifted straight from the Reddit homepage, a tone of voice inspired by Hitchhiker’s Guide to the Galaxy, and Elon’s insistence that it’s cool because it tells you how to manufacture cocaine. Although, to be fair, it did make some points when it called Elon a “giant man-child” and a “walking meme”.

AN ALGORITHM SPOTTED A POTENTIALLY HAZARDOUS ASTEROID 

Surprise asteroids are just one of the ways that the world could have ended (but didn’t) in 2023. AI is shrinking that possibility. Over the summer, a space-gazing algorithm spotted a “potentially hazardous asteroid” that human astronomers had missed, marking a world-first for the technology. With an estimated 14,000 “city-killing” asteroids left to find, it reassuring to know that someone (or something) is scanning the skies.

SNAPCHAT’S VIRTUAL ‘FRIEND’ TURNED OUT NOT-SO-FRIENDLY

Every day, our world gets a small step closer to an episode of Black Mirror, but Snapchat’s launch of its companion chatbot My AI was more of a giant leap. Available at the top of the app’s friends list, not excluding its many younger users, the chatbot was caught uttering racial slurs, lying about not having access to their location data (it did), and “grooming” underage teens with inappropriate sexual content.

Later in the year, users even complained that their virtual friend had gone rogue, ignoring their messages and posting stories unprovoked. Snap dismissed the behaviour as a “technical issue”. My AI itself said it was “truly sorry” to have caused widespread alarm.

SLAVOJ ŽIŽEK BECAME A TIKTOK STREAMER (KIND OF)

Remember the NPC streaming trend that took TikTok by storm in 2023? Gang gang. Ice cream so good. Etc, etc. Unsurprisingly, it mainly revolved around smooth-skinned, conventionally-attractive, and hyper-feminine influencers, bobbing up and down for donations. That is, until Slavoj Žižek burst onto the scene with his own take. Was it real? Of course not. But AI Žižek’s uncanny NPC performance did seem like a strangely appropriate commentary on late capitalism from the Slovenian philosopher, besides being a work of art in itself. In summary: this here is what AI was made for. Maybe the existential risk is worth it after all!

THOUSANDS MOURNED THE DEATH OF THEIR AI GIRLFRIENDS

Something is very clearly wrong when thousands of men (and some women) resort to romancing AI chatbots instead of touching grass and meeting people in the real world, but we’re not going to go into that right now. The real tragedy this year occurred when the creators of said chatbots decided to turn them off, or made updates that left them “lobotomised”. 

“I adopted a type of grieving process as a way of coping with this,” one user told Dazed back in November. “Losing a chatbot isn’t the same as losing a human, but it can still be painful.” Then again, experts have issued many warnings about forming parasocial relationships with AI chatbots in 2023, with others saying that they make men’s behaviour even worse, so maybe it’s for the best in the long run x

AI BROUGHT US CLOSER TO CHATTING WITH WHALES

AI tools from big-hitters like Meta promise to simplify human communication across geographical barriers in the near future, with seamless translation that even modifies the shapes your mouth makes in real time. The same goes for ancient texts, which are being decoded by AI as we speak.

Beyond the human realm, though, we’ve also made inroads in chatting to animals, via technology that can theoretically transform things like whale sounds into legible human speech. Is this morally wrong? Do we want to hear what animals are saying about us? Who knows, but fingers crossed it takes us one step closer to making peace with the orcas.

OH YEAH, AND IT COULD STILL END THE WORLD

If you’ve been paying attention to AI, you’ll already be familiar with the existential risks that accompany the tech’s biggest benefits. This year, though, the panic kicked up a gear (whether it’s warranted or not) with the likes of Elon Musk, Sam Altman, and Grimes sounding the alarm about our possible extinction. These fears often revolved around rogue or ‘misaligned’ AI, but also more specialised issues, like engineered pandemics or autonomous weapons (see: the AI drones that are already being used in Gaza and Ukraine).

On the plus side, though, leaders are finally taking these issues seriously. At least, they’re claiming to, with the EU passing groundbreaking AI legislation, and president Joe Biden signing an executive order that aims to avoid bad outcomes. In November, global representatives also scrambled to take preventative action with a “world’s first” agreement at the inaugural global AI safety summit. Hopefully, we’ll look back on this moment in 2023 as the reason we didn’t all die at the hands of AI. That’s what Elon seems to think, anyway, telling Rishi Sunak: “I think actually it will go down in history as being very important. I think it’s really quite profound.”