Meta is following in Google's footsteps in limiting the use of AI chatbots to avoid misleading or false information that might confuse voters. This time, Facebook's parent firm is bringing this policy to the ongoing general elections in India.
Restrictions on Election-Related Queries
(Photo: Muhammad Asyfaul from Unsplash) With India having the world's biggest election, Meta assures that any election-related queries on its AI chatbot will not interfere with people's perspectives on their chosen candidates.
Acknowledging concerns regarding the potential dissemination of misleading information during elections, Meta has confirmed its decision to limit specific election-related keywords for its AI chatbot during the testing phase.
This only shows that the company is committed to enhancing the AI response system to mitigate any misinformation risks, especially during this crucial time when a single misinformation can influence the perspective of the whole crowd.
"This is a new technology, and it may not always return the response we intend, which is the same for all generative AI systems. Since we launched, we've constantly released updates and improvements to our models, and we're continuing to work on making them better," a company spokesperson told TechCrunch.
Related Article : Meta Unveils Llama 3-Powered Meta AI Assistant
Tech Giants Regulate AI Responses During Election
As mentioned, the search engine giant has implemented similar restrictions on election-related queries through its Gemini chatbot in India and other election-bound markets.
Google believes that inaccurate responses by the AI chatbot could spell doom for the whole political scene. In just one click, if the response is detected to pose misleading details, it could mean "fraud" for the entire nation.
In addition to filtering AI responses, Meta has announced comprehensive measures to regulate its platform's content during election periods. This includes suspending political advertisements leading up to elections worldwide and identifying AI-generated content to ensure transparency.
Approach to Handling GenAI Queries
Meta's strategy involves directing users to authoritative sources, such as the Election Commission's website, when specific political figures or terms are queried.
By employing a blocklist mechanism, the social media titan aims to guide users toward reliable information sources while limiting the potential spread of misinformation.
While Meta's AI chatbot demonstrates proficiency in certain areas, inconsistencies have been noted. Despite redirecting users to the Election Commission's website for political queries, the chatbot's responses may vary based on the specific terms used, indicating room for improvement in its functionality.
Expansion Plans and Testing Phases
Although Meta has introduced its Llama 3-powered Meta AI chatbot in multiple countries, including the United States, its deployment in India remains in the testing phase.
The company announced its ongoing learning process, which uses user feedback and iterative testing methodologies to refine AI products and features.
In other news, Tech Times reported that AI image generation was now possible for WhatsApp users in the US. However, Meta clarified that it was still in the beta phase, so it's still not free of bugs and other uncertainties.
The real-time image generation feature is also available on Facebook, Messenger, and Instagram. Users can make GIFs out of their friends' pictures so they can share them on the platform.
