BOSTON (AP) – When US Rep. Jake Auchincloss decided to deliver a speech on the Bill that would create a US-Israeli artificial intelligence center, he chose to let AI do the talking.
The brief two-paragraph speech the Massachusetts Democrat read on the floor of the US House on Wednesday was created by online AI chatbot ChatGPT. Staff said they believe this is the first AI-written speech read in Congress.
Auchincloss said he asked the system to “write 100 words to send on the floor of the House of Representatives” about the legislation. Auchincloss said he had to refine the prompts several times to produce the text he finally read.
The bill, billed as Auchincloss, would establish the US-Israel AI Center in the United States to serve as a hub for AI research and development in the public, private and educational sectors.
Auchincloss said that part of the decision to read the text created by ChatGPT was to help debate AI and the challenges and opportunities it creates. He said he didn’t want to see a repeat of the social media incident, which started small and ballooned faster than Congress could handle.
“I’m the youngest parent in the Democratic caucus, AI will be a part of my life and it can be a general purpose technology for my children,” Auchincloss, 34, said.
The release of ChatGPT and other AI programs available on the internet has become a challenge for teachers who now have to contend with the possibility of students submitting AI-generated essays.
Researchers also worry that AI chatbots could help turbocharge efforts to spread misinformation and propaganda.
OpenAI, the non-profit organization that created ChatGPT, has admitted on its website that ChatGPT “can occasionally produce wrong answers” and that its responses will sometimes be misleading as a result of the way it learns. Users are advised to check whether the response is accurate or not,
The resulting text of the Auchincloss directive includes sentences like: “We must collaborate with international partners like the government of Israel to ensure that the United States maintains a leadership role in AI research and development and responsibly explores the many possibilities available to emerging technologies.”
“There were probably about a dozen of my friends on the floor. I’m sure none of them knew that it was written by a computer,” he said.
Lawmakers and others shouldn’t reflexively oppose new technologies, but they also shouldn’t wait too long before creating new policies or laws to help regulate them, Auchincloss said.
In particular, he said the country needs a “public balancer” for big tech companies that would help ensure that smaller developers and universities have access to the same cloud computing, advanced algorithms and raw data as big companies.