It’s now, more than ever, inevitable to come across Artificial Intelligence (AI) when we scroll on our phones and search on the internet.
Whether it’s the helpfulness of the built-in AI overviews within our Google searches, the ridiculous SORA AI videos we see on TikTok, emulating real life, or sneakily using ChatGPT for school help or as a friend, AI has infiltrated our lives and is here to stay.
Although AI can aid in general productivity, there are major concerns that have been raised in academia about its validity and the academic integrity of works using it. This automates the art of research, eliminating the process and lessons learned within it, and puts many careers in harm’s way. Besides academics, chaos has erupted in the world of the arts, raising questions about how long it will be until AI invades the art realm and further takes away career opportunities, along with human emotions conveyed in art.
One of these invasions of the arts world, specifically acting, is the creation of ‘Tilly Norwood’, an AI actress.
To better understand the creation of Tilly Norwood, we need to understand AI as a baseline. According to an essay by Michael Townsen Hicks of the University of Glasgow, “ChatGPT Is B*llsh*t”, AI is a ‘Large Language Model’ (LLM). This program carries the goal of simply mirroring the behaviours and answers of “human speech or writing”.
These LLMs are able to produce these human-like answers by information and patterns fed to the program by their creators, typically very selective and prone to biases. This information is fed in large quantities, and the LLM takes specific words to create webs of related words within its context, and picks up on patterns, which display as sets of words which may come before or after the given word.
In this way, AI is tested by researchers to answer questions, receiving negative or positive feedback on its responses, which enables better future communication and constant development. Whether this information and content being fed to the AI is text, facts, photos or videos, AI analyzes the humanism of these texts, and therefore can produce the most human-like response. With a field such as the arts, relying on the translation of these emotions and human conditions into physical visuals, acts or words, AI begins to devalue and generalize the very individual experiences of artists in all forms.
This aim to act ‘human-like’ is not only seen in AI-formed text, but also in a virtual individual. Tilly Norwood is an AI creation in association with Particle6, an AI company. Tilly Norwood takes on the appearance of a young-adult woman with a social media presence on Instagram. On her Instagram, consisting of 61.8K followers (as of October 11th, 2025), her bio reads: “You’ll either get it or pretend you don’t. I’m a creation. #aiart”.
Through posts showcasing her seemingly ‘everyday life’, including photos at cafes, professional headshots, and multiple videos of her in acting settings or roles, the identity of an actress is built. From a post emulating Sydney Sweeney’s American Eagle ad to a compilation of varying scenes, from horror to drama, she builds herself a portfolio of sorts online. Tilly Norwood, as well, addresses backlash in a post made on September 28th, stating that “ …she is not a replacement for a human being, but a creative work – a piece of art.”. This personification of AI into a person and lifestyle details built around her troubles many.
In response, many actors have stated their concerns about the dehumanization of the arts, losing its most valuable component: human emotion. The Screen Actors Guild – American Federation of Television and Radio Artists (SAG–AFTRA) has released multiple statements regarding the use of AI actors. Clarifying on their Instagram that, “‘Tilly Norwood’ is not an actor, it’s a character …”, labelling the AI actress as simply a ‘synthetic performer’, which is banned for producers’ use in its contracts. This proactive decision to announce and address such issues, protecting the acting field, exemplifies that problems are more effectively tackled when they are quantified.
Now, although AI can consume a plethora of knowledge and experiences to produce its own fused version of them, we must take into consideration the dangers this can have on the acting community. As characterized in the process and functions of an LLM, they take existing information and create responses to mimic these, meaning only one thing: Tilly Norwood’s AI was created through the viewing and note-taking of former scenes and roles of actors. This use of existing scenes, although open to the public, caused problems, as these actors were used without credit and not compensated for their forced role as a ‘teacher’ towards these LLMs.
This acknowledgement of acting observed also brings in the question of the mechanization of everything in our urban world. With the implementation of machines to take over tasks, sometimes just minimum wage to the average worker, such as ‘self check-outs’, the same questions are raised with this creation: is it simply made to catalyze cost cuts and take away the liabilities associated with humans?
Theoretically, an AI actor or worker can work overtime without significant pay, forgo benefits and insurance, and perform tasks that would require actors to take multiple takes or undergo extensive training. There is a grim possibility that companies might take advantage of this. Additionally, AI has no agency; no complaints can be made about comfort level with specific scenes, and therefore, can easily lead to taking advantage of this actor in roles typically not considered by others after thorough thought and informed consent. These components of monetary value lead to this possible use of the AI as money-centred and greedy, rather than an art piece or advancement.
Although the technological world has created, developed and used AI to facilitate advancements in the academic world, and in social media, its overuse in human and emotion-reliant fields can lead to a dystopian reality of detached art.