According to a recently published study by Meta and Stanford University, participants in a forum on artificial intelligence (AI) hosted by Meta had a more positive perception of AI's potential impact. As a result, Meta plans to organize similar forums in the future. The findings were part of a study conducted on a forum where participants received information about AI from experts, academics, and other stakeholders. They also discussed policy proposals regarding AI chatbots. The study, presented at the October 2023 Meta Community Forum by Stanford's Deliberative Democracy Lab, revealed that initially, 49. 8% of the 393 American participants believed that AI had a positive impact. However, after participating in the forum, this percentage increased to 54. 4%, indicating a 4. 6% increase in positive perceptions of AI's impact. In addition to American participants, the forum included participants from Brazil, Germany, and Spain. Participants from these countries already held a larger majority with a positive view of AI, and their positive opinions were further reinforced during the forum. More participants from these countries had also previously used AI chatbots like ChatGPT compared to American participants. Meta, the company that owns platforms like Facebook, Instagram, and WhatsApp, has already launched generative AI products such as Imagine, which can generate images based on textual prompts. The company has also integrated Meta AI chatbots into some of its apps, with some featuring the faces of celebrities like TikTok influencer Charli D’Amelio. Meta, however, is also aware of the potential negative aspects of generative AI. In March, NBC News reported that multiple ads had run on Meta's platforms for an AI-powered deepfake app that claimed to "undress" pictures of women and girls.

One of the ads featured an image of actress Jenna Ortega when she was 16. These ads appeared at a time when students in middle and high schools across the U. S. were found creating sexually explicit deepfake images of their classmates. Previously, NBC News had also discovered ads for a similar deepfake app featuring fake sexually suggestive videos of actress Emma Watson. After being contacted by NBC News, Meta suspended both deepfake apps from advertising. The AI forum hosted by Meta and Stanford focused on exploring how generative AI, including chatbots, should interact with users as the technology advances. Thirty-eight policy proposals were discussed in small groups, allowing participants to formulate questions for experts. An "Advisory Committee" consisting of distinguished members chose the experts, but their identities were not disclosed. During the forum, participants deliberated on various questions, such as whether AI chatbots should be capable of forming romantic relationships with humans, which sources of information they should rely on, whether they should exhibit human-like behavior, whether they should be allowed to be offensive, and how much transparency should be provided to users regarding their artificial nature. In their conclusions, Meta and Stanford highlighted that participants expressed concerns regarding AI bias, misinformation, and potential violations of human rights. Participants also emphasized the importance of having control over chatbots' access to their data, skepticism towards chatbots replacing human interaction, and the need for tech companies to prioritize user privacy, data security, and transparency in data usage.