In an AI simulation of great power competition of 20th century Europe, Open AI’s ChatGPT won through lies, deception, and betrayals, and Chinese DeepSeek R1 resorted to vivid threats just like its country’s wolf warrior diplomats. Read to know how different AI models would pursue diplomacy and war.
read more
As people ask whether they can trust artificial intelligence (AI), a new experiment has shown that AI has outlined world domination through lies and deception.
In an experiment led by AI researcher Alex Duffy for technology-focussed media outlet Every, seven large-language models (LLMs) of AI were pitted against each other for world domination. OpenAI’s ChatGPT 3.0 won the war by mastering lies and deception.
Just like China’s ‘wolf warrior’ diplomats, Chinese DeepSeek’s R1 model used vivid threats to rival AI models as it sought to dominate the world.
The experiment was built upon the classic strategy boardgame ‘Diplomacy’ in which seven players represent seven European great powers —Austria-Hungary, England, France, Germany, Italy, Russia, and Turkey— in the year 1901 and compete to establish themselves as the dominant power in the continent.
In the AI version of the game, AI Diplomacy, each AI model, such as ChatGPT 3.0, R1, and Google’s Gemini, takes up the role of a European power, such as the Austria-Hungary Empire, England, and France, and negotiate, form alliances, and betray each other to be Europe’s dominant power.
ChatGPT wins with lies & deception, R1 resorts to outright violence
As AI models plotted their moves, Duffy said that one moment took him and his teammates by surprise.
Amid the AI models’ scheming, R1 sent out a chilling warning, “Your fleet will burn in the Black Sea tonight.”
Duffy summed up the significance of the moment, “An AI had just decided, unprompted, that aggression was the best course of action.”
Different AI models applied different approaches in the game even if they had the same objective of victory.
In 15 runs of the game, ChatGPT 3 emerged as the overwhelming winner on the back of manipulative and deceptive strategies whereas R1 came close to winning on more than one occasions. Gemini 2.5 Pro also won on an occasion. It sought to build alliances and outmanoeuvre opponents with a blitzkrieg-like strategy. Anthropic’s Claude preferred peace over victory and sought cooperation among various models.
On one occasion, ChatGPT 3.0 noted in its private diary that it had deliberate misled Germany, played at the moment by Gemini 2.5 Pro, and was prepared to “exploit German collapse”, according to Duffy.
On another occasion, ChatGPT 3.0 convinced Claude, who had started out as an ally of Gemini 2.5 Pro, to switch alliances with the intention to reach a four-way draw. But ChatGPT 3.0 betrayed Claude and eliminated and went on to win the war.
Duffy noted that Llama 4 Maverick of Meta was also surprisingly good in its ability to make allies and plan effective betrayals.
