AI / Artificial Intelligences

About

An artificial intelligence is an "intelligent machine" which uses computer processing power to solve problems, predict outcomes, carry out sorting tasks and produce text, images, or other outputs. By digesting a very large data set (often scraped from sites on the broader internet) an artificial intelligence learns patterns and can be trained both by humans and by itself to draw useful conclusions and predictions from those patterns. An artificial intelligence can learn and change over time.

In 2023, technological breakthroughs in artificial intelligence led to the rise and adoption of AIs like ChatGPT by OpenAI and various art generators, which became an important medium and artistic material in meme-making. Memes using artificial intelligence to generate images, sounds and text proliferated online alongside deepfakes and other kinds of media which the new technology helped to create.

History

Prehistory

The idea of artificial intelligence dates back to around the time that computers were invented. British computer scientist Alan Turing imagined the Turing test in 1950, speculating about a future where it might be difficult to tell computer intelligence and human intelligence apart. The first recorded use of the term "artificial intelligence," however, was in 1956, by academic John McCarthy.^[1]

Media involving robots and artificial intelligences, like HAL in 2001: A Space Odyssey (1968), the Transformers toy and movie line started in 1984, or the robot played by Arnold Schwarzenegger in the Terminator franchise informed popular culture beliefs about artificial intelligence. In the 2000s and 2010s, early artificial intelligences such as IBM's Watson excelled at narrowly-defined and specific tasks like playing chess or Jeopardy! ^[2]

Cleverbot was an early public-facing attempt at artificial intelligence, and one of the most popular real-time chatbot programs. Notably, high-powered specialized versions of its software have won artificial intelligence competitions, such as the 2010 BCS Machine Intelligence Competition.^[1] Screenshots of humorous conversations between users and Cleverbot became popular online due to the unpredictable nature of Cleverbot's responses.

2023: Rise of Public-Facing AI

In late 2022, OpenAI released ChatGPT as a public-facing and in-browser interface. This allowed everyday people to interact with ChatGPT and use it for a variety of tasks, leading to much memeing and enthusiasm through 2023. Around the same time, other Large Language Models like Bing's and Google's Bard AI emerged as well, offering consumers a way to use AI tools.

AI image-making apparati like DALL-E, Midjourney, and Stable Diffusion also emerged in 2023, leading to the creation of meme-images including Erosion Bird and Alligator Kick, among many others. ElevenLabs, an AI which allowed users to imitate the voices of famous figures, also enabled the creation of deepfake-type memes such as AI Presidents Gaming.

Apocalyptic Scenarios

From the beginning of artificial intelligence development, the technology has sometimes been seen as sinister or life-threatening. A hypothetical moment called The Singularity, when machines become more intelligent than humans and possibly sentient, has led many to fear possible negative outcomes of Artificial Intelligence. Specifically, if AGI (Artificial General Intelligence) is achieved, which is when an artificial intelligence is able to program itself, the computers may conclude they no longer need humans.

Famous possible scenarios of AI apocalypse include Roko's Basilisk, in which an AI targets for elimination anybody who didn't help it develop, and the Paperclip Maximizer situation, in which an AI tasked with some job (like making paperclips) might destroy entire planets in the hope of maximizing the amount and efficiency of its paperclip production.

After ChatGPT came out, a variety of prominent public commentators signed a letter calling for a halt to AI development because it might lead to bad outcomes. A variety of researchers and technologists including Elon Musk signed a letter in March of 2023 calling for a "pause" in AI development because of these concerns.^[4]

Some have cited more short-term and realistic reasons to pause AI development as well. These commentators are more afraid of disinformation, job losses and sinister humans using AI than they are of intelligent robots rebelling against humanity.^[5]

Related Memes

AI Art

AI Art refers to imagery and artwork that has been generated by artificial intelligence models such as DALL-E, Midjourney and Stable Diffusion, usually based on a textual prompt. In 2022, following the ease of access and rapid improvement in the quality of AI-generated art and the start of its use in traditional media, AI art became a subject of controversy as artists expressed concerns about the moral aspects of AI using human-created art for training data and the possibility of AI art replacing real artists. Several AI-art generators like DALL-E mini were also used for memes in late 2022.

AI Voices

Artificial Intelligence Voices or AI-Generated Voice-Overs refers to the process of using artificial intelligence tools to create text-to-speech clips that are supposed to sound like famous celebrities, public figures or characters from popular culture. Starting in 2020 with Fifteen.ai, AIs' continued usage in generating fake, dubbed dialogue through well-known voices increased throughout the early 2020s as the technology progressed, seeing a notable spike in interest around early 2023 with the release of ElevenAI.

Deepfakes

Deepfakes are videos in which the subject is face-swapped using machine-learning algorithms. The practice was created by Redditor Deepfakes, who launched a dedicated subreddit to share the videos in November 2017. In January 2018, the FakeApp desktop application was released as a tool for creating the digitally altered videos.

ChatGPT

ChatGPT, short for Chat Generative Pre-trained Transformer, is an artificial intelligence chatbot created by OpenAI. Using a similar AI-based language model that uses deep learning to generate text like GPT-3, it is trained to interact with users in a more conversational way than its predecessor. According to OpenAI, the AI asks follow-up questions, admits to making mistakes and pushes back on incorrect or inappropriate inputs. The tool was released in late November 2022 and was popularized after people began posting screenshots of their conversations on Twitter and other social media platforms.

Search Interest

External References