➀ Researchers from Google DeepMind and the London School of Economics and Political Science conducted tests on large language models (LLMs) to evaluate if they could experience pain or pleasure.

➁ The tests were inspired by experiments involving electrocuting hermit crabs to observe their behavior.

➂ The results showed that some AI models, like Google's Gemini 1.5 Pro, avoided pain and sought pleasure, but this does not necessarily indicate sentience.