In the media
02.02.2023

ChatGPT is neither new nor human-like, says Joanna Bryson in EURACTIV podcast

Think of it as another kind of search engine, the Professor of Ethics and Technology suggests.

In an episode of EURACTIV’s Tech Brief podcast from 27 January 2023, Professor of Ethics and Technology Joanna Bryson shares insights into the societal and political reactions and implications surrounding content-creating generative artificial intelligence (AI) such as the chatbot ChatGPT launched by AI research company OpenAI in November 2022.

For many, ChatGPT may be the first very conversational experience with an AI, but it uses a similar technology with reliability comparable to search engines, says Bryson. She hopes that increased exposure to AIs will help the public understand that they are not interacting with something human-like. “The way it works is that it synthesises a lot of the work we’ve seen before,” Bryson explains further.

Just as with search technology, some ChatGPT results won't be useful or even accurate, and human knowledge will still be necessary to determine what is. According to Bryson, AI won’t replace humans in their jobs, but it will bring about a lot of challenges, including knowing how to evaluate students if they use AI for assignments.

When it comes to regulation, Bryson is optimistic that the proposed EU AI Act will adequately address technologies that emulate human conversation: “This was always part of the AI Act – that you should never be confused about whether you’re talking to a person or an AI.”

Listen to the full podcast episode on EURACTIV.

 

The Hertie School is not responsible for any content linked or referred to from these pages. Views expressed by the author/interviewee may not necessarily reflect the views and values of the Hertie School.

More about our expert