HomeNewsPerplexity launches an election tracker

Perplexity launches an election tracker

Perplexity, the AI-powered search engine, is perhaps hallucinating every now and then. However, the corporate wants to indicate that it’s trustworthy enough to trace election results.

Today, helplessness announced a dedicated hub for US general election information. Based on data from The Associated Press and Democracy Works, the corporate described this in a Blog as “an entry point to understanding key issues.”

“Starting Tuesday, we are going to offer live election updates using data from The Associated Press to maintain you informed about presidential, Senate and House races on the state and national levels,” Perplexity wrote. “Thank you Democracy Works for giving us access to your election API to enable these experiences.”

Perplexity didn’t immediately reply to TechCrunch's inquiry about whether the corporate compensates AP and Democracy Works for this data.

Photo credit:confusion

Perplexity's Election Hub answers election-related questions equivalent to voting requirements and poll times, in addition to AI-aggregated evaluation on voting measures, candidates, policy positions and endorsements. Summaries aside, the hub is basically a wrapper for data from The Associated Press and Democracy Works API. What's notable, nonetheless, is that Perplexity's competitors have shown reluctance to introduce comparable features as a consequence of fears of AI-generated misinformation.

In its recently released ChatGPT search experience, OpenAI redirects users asking about election results to The Associated Press and Reuters. Anthropic's Claude chatbot doesn't answer questions on election results, and neither does Google's Gemini.

Given AI's poor track record on this area, we will only hope that Perplexity's hub is as accurate as the corporate claims.

In a July studyThe Center for Democracy and Technology found that across 77 different election-related queries, greater than a 3rd of responses generated by AI chatbots, including Claude and Gemini, contained false information. Other research has shown that enormous chatbots perform even worse when asked questions on elections Accessibility challenges and people whose primary language isn’t English.

LEAVE A REPLY

Please enter your comment!
Please enter your name here

Must Read