When AI first appeared, it raised new ethical questions. Either a tool to improve our workflow or an opportunity for big companies to make savings on the workforce, it is hard to understand the impact Artificial Intelligence will have on our daily life, workplace and opportunities. If you find yourself wondering what this is all about and how it could affect you, this article might help you find your stance.
A powerful tool
The term “Artificial Intelligence” was first coined in 1956 by scientist John McCarthy. It relied on the premise that “every aspect of learning or any other feature of intelligence can in principle be so precisely described that a machine can be made to simulate it”¹. Since then, “deep learning” techniques², allowing computers to learn using data, have been developed. And as computers became more performant and popular, data grew infinitely bigger. And so did AI.
The first notable breakthrough happened in 1997 in the field of chess, when the engine “Deep Blue” first defeated then World Chess Champion Garry Kasparov. More recently, AI became a hot topic again when OpenAi launched ChatGPT, a language model that enabled users to give instructions or ask questions through text input. AI is a fast-evolving technology that has many uses, from Chatbots on Google or Snapchat to image-generating AIs such as DALL·E 2, Midjourney or Stable Diffusion, which allow users to create all sorts of visuals using text prompts. Even the field of architecture has been affected, with plan generators Maket.AI and ARCHITEChTURES.
Is AI ethical ?
Very quickly, soaring concerns arose about the unethical use of data made by AI, as well as its general use :
- AI enables people to create images, audio and video content, using the images and voices of public figures saying or doing things they haven’t actually said or done. These are called Deep Fakes and they create risks³ of disinformation and potentially traumatic experiences. For example, they can alter people’s opinions on a politician, influence votes or be used as scapegoats by someone refusing to take accountability, giving them the perfect argument for denying the validity of a real video. In that way, AI can become a tool of disinformation. Deep Fakes can also be psychologically disturbing to people seeing themselves do or say things they never have said or done. Using AI portrait generators, anyone can create fake social media profiles with faces of people who aren’t actually real. Even more worrisome, a study⁴ has found that not only are we unable to tell the difference between AI-generated faces and real ones, we also trust AI-generated faces more than the faces of real people.
- Artificial Intelligence questions the very nature of art, too: in the Netherlands, The Mauritshuis museum exhibited an AI-generated image of ‘Girl with a pearl earring’⁵. Last August, an art competition awarded its first place and a 300$ prize to an AI-generated image “Théâtre D'opéra Spatial” (see header image), which was later denied any copyright protection by the US Copyright Office⁶ for “not being a product of human authorship” .
- Between May and September 2023, screenwriters in the US were on strike, demanding better compensation⁷ and stricter rules around the subject of AI. The place of AI in the film and TV industry was part of the demands of the writers’ union⁸, as they didn’t want their copyrighted work to be used to train AI.
In cases like these, as is the case with many new technologies, it is hard to get legal support. Current regulations are falling behind and need to be adapted.
Current state of regulations
AI relies on data on the internet. Following this logic, the quality of an AI program's output relies on the quality and quantity of its database – and this database can be subject to copyright. Because, until now, everything ever published on the internet (articles, images, videos) has been made by humans and can be subject to copyright.
The regulation around the use of and acces to this data is an ongoing process. Some companies have already taken action: X (formerly known as Twitter) announced in July 2023 a restriction on the number of tweets available depending on the type of account. New unverified accounts, commonly used for data scraping for AI training, were restricted to reading 500 tweets daily⁹. In the same month, Reddit started charging developers for using their API¹⁰ (Application Programming Interface, a way to access their data), in an attempt to stop companies from using the platform's data for AI training. The European Guild for Artificial Intelligence Regulation (EGAIR) has been working on getting these uses regulated, resulting in the AI act being passed in May 2023¹¹. The AI act aims to make the foundation models (the data used to train the program) transparent as well as to defend and support the privacy rights of European citizens in regards to their data.
As far as copyrights, some regulations are starting to get passed as the US Copyright Office took its stance by rejecting copyright for AI generated images, as it lacked human authorship¹². The EU stance is less straightforward as “Much of the literature on AI and copyright focuses on the scenario of the AI system producing content with only limited input on the part of the user of the system. If the role of the user of the system is indeed so constricted that they cannot exercise free choice at any stage of the creative process, the user will not qualify as author of the ensuing production”¹³ . How much "free choice" in the creation process is required to qualify for authorship is still unclear, so in the meantime, the decision is left to each country to make.
AI can be a powerful tool for the workers of the future, making some tasks more efficient and creating some well-deserved free time; but if we don't pay attention, it can also turn against us and become another minion of the capitalist machine.
What are some questions you have about the impact of AI on your life?
Sources:
¹ “Artificial Intelligence Coined at Dartmouth”, Darthmouth
² FOOTE, Keith D. “A Brief History of Deep Learning”. Dataversity, 04 Feb. 2022,
³ SAMPLE, Ian. “ What are deepfakes – and how can you spot them?”. The Guardian, 13 Jan. 2020
⁴ MASTERSON, Victoria. “People trust AI fake faces more than real ones, according to a new study”. World Economic Forum, 15 Mar. 2022
⁵ HARRIS, Gareth. “ Online storm erupts over AI work in Dutch museum’s ‘Girl with a Pearl Earring’ display. The Art Newspaper, 13 Mar. 2023
⁶ EDWARDS, Benj. “ US rejects AI copyright for famous state fair-winning Midjourney art". Ars Technica, 09 Nov. 2023
⁷ LOWE, Lindsay. “ The 2023 writer’s strike explained - and how it will affect your favorite TV shows”. Today, 10 May 2023
⁸ WALFISZ, Jonny. “Why the AI demand from the Writers Guild strikes is the most important talking point”. Euro news, 03 May 2023
⁹CURSINO Malu, CLAYTON James, “Twitter temporarily restricts tweets users can see, Elon Musk announces”. BBC News, 01 Jul. 2023
¹⁰LUTKEVICH, Ben, “Reddit pricing: API charge explained”. TechTarget, 11 Jul. 2023,
¹¹ “New AI Act voted and approved by the Parliamentary committee - A first step in the right direction”. EGAIR
¹² BRITTAIN, Blake. “AI-generated art cannot receive copyrights, US court says”. Reuters, 21 Aug. 2023
¹³ HUGENHOLTZ, P. Bernt and QUINTAIS, João Pedro. “Copyright and Artificial Creation: Does EU Copyright Law Protect AI-Assisted Output?”. Springer Link, 04 Oct. 2021