SNL Papyrus Avatar Sketch Reminds of Current AI Weaknesses

By Todd Russell Feb5,2024

Recently watched the 2017 SNL sketch below (and yeah, how has it been 6+ years and haven’t seen this yet) and it’s pure LOL. If you haven’t seen yet, it’s up there with Christopher Walken’s Blue Oyster Cult “more cowbell.”

Apparently, Avatar creator James Cameron and his team thought about making changes, seriously, to the Avatar branding during the making of the Avatar sequel and ultimately decided to leave it alone.

While speaking to People about Ryan Gosling’s comedic bit, the Oscar-winning director shared that he thought “the font thing’s funny” before sharing that the sketch did lead to a conversation behind the scenes regarding whether Avatar: The Way of Water should lean into its branding or go into another direction. Today, the font is mostly the same, except it’s somewhat bolder than the OG design. Back in 2020, people noticed the change to the logo for the first Avatar movie on Disney+, which somewhat strayed away from Papyrus. You can check out the sketch that arguably initiated all of this down below:

James Cameron Humorously Recalls How Ryan Gosling’s Avatar Font Jab In SNL Sketch Created A ‘Dilemma’ For The Way Of Water | Cinemablend

On December 7, 2023 the movie tie-in game Avatar: Frontiers of Pandora released and I checked it out, streaming some of the first thoughts gameplay after playing on Ubisoft+ which, in my research, was the least expensive way to try out the game (see: Ubisoft+ time again December 7 for Avatar: Frontiers of Pandora).

The SNL video is so good that it makes me think of the logo/branding now every time it appears. Ironic how it talks about branding while mocking the branding. Font choices are the kinds of things graphic designers do care deeply about. It’s just one more part of the art which also makes me think of AI. When will AI have that same level of passion? Will it ever in my lifetime?

Recently, on the Twitch stream I’ve been playing around with an AI co-host, that I call “Clone Todd” which is essentially my voice cloned with AI. The voice talks too fast and doesn’t have the enunciation of words quite like I speak, but at times it’s eerily close. Several folks have commented that it’s like a younger voice of mine, but it’s just way too fast and the rhythm is off. The passion just isn’t there. Eventually I added more clones of some other people online using voice samples from their YouTube videos. These clones actually sound a little more realistic than mine — sometimes.

This, to me, is the single reason at least in creative applications (and let’s use games for example) AI is going to be slow in completely taking over for human beings. AI is already being used, but it’s noticeable and that will turn people off from buying and playing games with excess AI – at least initially.

Already there are some businesses using AI customer service bots with dumb results:

A seemingly fed-up Beauchamp then decided to see what the bot would be able to do — and as it turns out, the AI proved much more adept at denigrating DPD and spouting profanities than it was at providing customer service.

AI Customer Service Bot Disabled After Trashing Company Using It (futurism.com)

In time, the AI will improve and someday become difficult, if not impossible to recognize, but I’m not sure that happens in my lifetime anyway.

Mark Zuckerberg, the man behind Facebook, after spending a ridiculous amount of money in a metaverse reality that nobody wanted has switched gears to building something eerily like Skynet from the Terminator movies.

AI and AGI are already very nebulous terms, but in a nutshell, with “general intelligence” systems Zuckerberg wants to create much, much smarter computing systems — ones that at least match human cognitive abilities like learning, reasoning, planning, creating and remembering information.

Dream On, Mark Zuckerberg. Your New AI Bet Is a Real Long Shot – CNET

Check out just how many super powerful AI computers Zuck is putting together. It’s a bit mind-bending and … disturbing.

But let’s keep it lighter. What about the games? AI in games is still under great scrutiny.

Recently, Valve has relaxed but clarified their rules regarding games containing AI, requiring developers to fill out a survey:

When developers fill out a survey for Valve to get their game on Steam, it now includes AI disclosures developers have to fill out and they separate it into two categories: [Pre-generated AI and Live-Generated]

Valve announces new rules for games with AI Content on Steam | GamingOnLinux

The fact that AI was more widely used in 2023, though, does bring a bit of chill to the bones. Just for full disclosure, the artwork used in this article at the top was created using generative AI. I have been using NightCafe.Studio in some graphic art situations and also to play around and see what the technology is capable of doing. That creative urge and curiosity of mine: testing and play around with things and see what/how/why/when/where they work/don’t in real world situations.

It’s this sort of seemingly innocent curiosity blending human-created with AI that help grow more widespread use of AI. It’s already happening on a wide scale.

The current most popular game on Steam, Palworld, is under fire for allegedly using AI to create Pokemon-like game assets, of which the developer to date has strongly denied.

A subset of artists, game developers and other concerned citizens have taken to accusing the game of two things, mainly. First, ripping off bits and pieces of Pokémon to assemble their “Pals,” but second, an often simultaneous accusation that Palworld developer Pocket Pair has used GenAI in the game.

‘Palworld’ Accused Of Being ‘An AI Product,’ With No Evidence So Far (forbes.com)

The biggest problem I see with current “AI” and put in double quotes intentionally is that it is build upon learning and using existing creative works. Take, for example, the art generated. It too often uses and blends styles of artists and AI generations can look like something the artist did, but usually with some obvious problems like AI’s current struggle using words as text. I have an article on 1980 that shows up as “19080” … not quite what I wanted, but left in, because it’s somewhat humorous and helps to show it wasn’t human-created.

Bottom line: gamers en masse currently do not seem ready to accept AI-generated assets in games, but how long will this resistance last? Do you care, friendly gamer reader, about AI assets used in games? Would you rather human beings did every single thing in games? Some form of AI has existed in games since their onset. Think enemy reactions.

We discussed AI in games on the S2E2 of High Noon with Hollywood Polo show. We covered how far we want AI to go in games to enemy logic. It won’t be as much fun or fair gameplay when AI goes too far, when an AI enemy is too powerful and can way out-think human beings.

For now, anyway, AI is a long ways off from precision. Decisions like what font to use for Avatar … I’m not sure AI would be any better than whomever decided to use the Papyrus font. You?

Spread the love

Related Post

Leave a Reply

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