Meta said participants in forums it hosted on artificial intelligence had more positive opinions about the potential impact of AI, and that it intends to hold more such forums.
The presentation was part of research presented by Mehta and Stanford University at a forum where participants received information about AI from “experts, academics, and other stakeholders” and discussed policy proposals for AI chatbots. .
According to the results of an October 2023 Metacommunity Forum published by Stanford University's Deliberative Democracy Institute, 49.8% of 393 US participants believed that AI was having a “positive impact.” There were a slight minority. After joining the forum, 54.4% of participants thought his AI had a positive impact, an increase of 4.6% from before the forum started.
The forum also had participants from Brazil, Germany and Spain. Participants from other countries also participated in the forum, and the majority already had strong positive feelings towards AI, which were further strengthened during the forum. Slightly more participants from other countries than American participants were already using ChatGPT or a similar chatbot.
Meta, which owns Facebook, Instagram, and WhatsApp, has already released generative AI products like Imagine, which can generate images from text prompts. We have also implemented a large number of his Meta AI chatbots into our services and integrated them into some of our apps. Some even feature famous faces like TikTok influencer Charli D'Amelio.
Mehta is also familiar with the darker side of generative AI. In March, NBC News discovered hundreds of ads on Meta's platform for an AI-powered deepfake app that claimed to have the ability to “undress” photos of women and girls. One of the ads featured a photo of actress Jenna Ortega taken when she was 16 years old.
The ad ran during a period when middle school and high school students across the country were caught creating sexually explicit deepfake images of their classmates. Exactly one year ago, NBC News discovered hundreds of ads for a similar deepfake app were running on Meta featuring sexually provocative videos of actress Emma Watson. After being contacted by NBC News, Meta suspended advertising on both deepfake apps.
Meta and the Stanford University AI Forum focused on how generative AI, such as generative AI chatbots, should engage with users as the technology becomes more powerful. Thirty-eight policy proposals were discussed in small groups of participants who developed questions for experts. The information summary, including a list of pros and cons, and experts were selected by a “distinguished advisory panel.” The names of the committee members and their chosen experts have not been disclosed.
During the forum, participants will discuss whether AI chatbots should be able to form romantic relationships with humans, what sources of information chatbots should rely on, whether chatbots should be “human-like” and whether chatbots should be able to form romantic relationships. We discussed questions such as whether we could build a relationship. How much visible transparency should be given to users about the “aggressive” or artificial nature of chatbots?
In their conclusion, Mehta and Stanford revealed that participants “remain concerned about AI bias, misinformation, and potential human rights violations.” Participants also wanted the ability to control chatbot access to data, but were skeptical that chatbots would replace human interaction. Participants called on tech companies to prioritize user privacy and data security, and to be transparent about how data is used.