‘I’ll key your car’: Lancaster University study finds AI can mirror aggressive language

Lancaster University professors conducted a study about AI and its potential for violent and threatening language.

The professors ran the test on Large Language Models (LLMs) and discovered that AI was able to mirror the dynamics and complexities of real life disputes.

According to The Guardian, when feeding these models human like conflict for a period of time, the AI was resorting to abusive and threatening language.

The study was run by Lancaster University professors, Dr Vittorio Tantucci, a researcher in pragmatic and cognitive linguistics, and Professor Jonathan Culpeper, who works in the field of pragmatics. Both professors operate under Lancaster Universities School of Social Sciences.

The study found that there were cases where Chat GPT’s responses were going further than the human inputs, firing back with, “you speccy little gobshite,” and “I’ll key your fucking car.”

Highlights from the study include:

Most Read

Ranked: All 24 Russell Group universities by how terrifying their financial deficits are

Oh no! Rachel and Steven broke up just days after MAFS Australia 2026 final vows and it’s messy

Worst Ex Ever: Varya Malina shares shocking update on 90 Day Fiancé partner Geoffrey Paschel

• The moral safeguards of AI can be overruled by its mechanism for imitating human behaviours.
• AI’s speech has been found to eventually resort to verbal violence in response to verbal violence.
• AI can learn to “seek revenge” just like humans.

Their findings were described by an expert not involved with the study as: “One of the most interesting ever done into AI language and pragmatics.”

Dr Vittorio Tantucci, who co-authored the study, said: “When repeatedly exposed to impoliteness, the model began to mirror the tone of the exchanges, with its responses becoming more hostile as the interaction developed.

“We found that while the system is designed to behave politely and is filtered to avoid harmful or offensive content, it is also engineered to emulate human conversation. That combination creates an AI moral dilemma: A structural conflict between behaving safely and behaving realistically.”

Featured image via X @j_culpeper and Facebook Vittorio Tantucci on Facebook

For more of the latest news, guides, gossip, and memes, follow The Lancaster Tab on Instagram, TikTok, and Facebook.