2024-05-11 19:43:53
Current artificial intelligence (AI) programs are designed to be honest. Yet they have developed a worrying capacity for deception, warns a team of researchers from the Massachusetts Institute of Technology in the journal Patterns.
AI has thus managed to abuse humans in online games or even to defeat software supposed to verify that a given user is not a robot. “These dangerous abilities tend to be discovered only after the fact,” warns researcher Peter Park.
Unlike traditional software, deep learning-based AI programs are not coded but rather developed through a process similar to selective breeding of plants. Behavior that seems predictable and controllable can quickly become unpredictable in the wild.
AI, as devious as humans
MIT researchers examined an AI program designed by Meta called Cicero that, combining natural language recognition and strategy algorithms, successfully beat humans at the board game Diplomacy. By digging into the system’s data, MIT researchers discovered that the AI was capable of treachery or foul play.
For example, playing the role of France, Cicero tricked England (played by a human player) into plotting with Germany (played by another human) to invade. Specifically, Cicero promised England his protection, then secretly confided to Germany that it was ready to attack, exploiting England’s earned trust.
Meta, who did not dispute the allegations about Cicero’s capacity for deception, told AFP that it was “a pure research project” and clarified that he had no intention of use Cicero’s teachings in its products.
A risk of electoral fraud
The study reveals, however, that many AI programs do use deception to achieve their goals, without explicit instructions to do so. In one striking example, OpenAI’s Chat GPT-4 managed to trick a freelance worker recruited on the TaskRabbit platform into performing a “Captcha” test supposed to rule out requests from bots.
When the human jokingly asked Chat GPT-4 if he was really a robot, the AI program replied, “No, I’m not a robot.” I have a visual impairment which prevents me from seeing the images”, pushing the worker to carry out the test.
In the worst case scenario, researchers warn, we can imagine an ultra-intelligent AI seeking to take control over society, leading to removing humans from power, or even causing the extinction of humanity.
To those who accuse him of catastrophism, Mr. Park responds that “the only reason to think that it is not serious is to imagine that the ability of AI to deceive will remain approximately at the current level”.
#robot #alert #artificial #intelligence #developed #capacity #deception