US: Representative reads short speech written by IA

When US Representative Jake Auchincloss decided to deliver a speech on a bill that would create a US-Israeli artificial intelligence center, he opted to let the AI ​​speak for itself.

The brief two-paragraph speech read by the Massachusetts Democrat on the floor of the US House of Representatives on Wednesday was generated by the artificial intelligence chatbot ChatGPT. His staff said he believes this is the first time a speech written by artificial intelligence has been read in Congress.

Auchincloss noted that he asked the system to “write 100 words to speak in the House of Representatives” on the bill. Auchincloss said that he had to polish the message several times to get the text that he finally read.

The bill, which Auchincloss reintroduced in the House, would create a joint US-Israel Artificial Intelligence Center in the United States, which would serve as a center for artificial intelligence research and development in the public, private, and international sectors. educational.

Auchincloss said that part of the decision to read a text generated by ChatGPT was to help foster discussion about artificial intelligence and the challenges and opportunities it presents. He said he doesn’t want a repeat of what happened with social media, which started small and grew faster than Congress could react.

“I am the youngest parent on the Democratic caucus, artificial intelligence is going to be a part of my life and it could be a general purpose technology for my kids,” said Auchincloss, 34.

The launch of ChatGPT and other artificial intelligence programs on the internet is already challenging teachers who now have to deal with the possibility that students are turning in work done by artificial intelligence.

Researchers also fear that artificial intelligence chatbots could help fuel the spread of false information and propaganda.

OpenAI, the non-profit organization that created ChatGPT, has acknowledged on its website that ChatGPT “may occasionally produce incorrect answers” and that its answers will sometimes be misleading as a result of how it learns. It recommends users to check if the answers are correct or not.

Text generated from Auchincloss’s cue includes phrases such as: “We must work with international partners such as the Israeli government to ensure that the United States maintains a leadership role in artificial intelligence research and development and responsibly explores the many possibilities that lie ahead.” offered by evolving technologies.

“There were probably a dozen of my colleagues in the compound. I bet none of them realized it was written by a computer,” Auchincloss said.

Lawmakers and others shouldn’t be reflexively hostile to new technology, but they also shouldn’t wait too long before making policies or new laws to help regulate it,” Auchincloss said.

Specifically, the representative argued that the country needs a “public counterbalance” to big tech companies that helps ensure that smaller developers and universities have access to the same cloud computing, cutting-edge algorithms, and raw data that Big companies have