At first glance, SearchGPT appears to be a carbon copy of Perplexity. The main differences are subtle: SearchGPT tends to display more images, and there are some minor design variations. However, both platforms could significantly improve their citation methods.
Curiously, both SearchGPT and Perplexity follow a similar pattern: they present a paragraph of information, followed by a list of references with links to sources at the end:
While this approach has its merits, a more granular citation system would be far more beneficial. Ideally, each specific statement would be linked directly to the relevant excerpt from the source material. Take Google’s Featured Snippets as an example. They use the #:~:text= parameter in URLs, making it easier for users to locate the exact information they’re looking for:
While this level of detail may not always be necessary, it would be incredibly useful in many cases. It would help users verify information and reduce concerns about potential hallucinations or inaccuracies introduced by the LLM.
By implementing more precise citation methods, SearchGPT could set itself apart and provide a more trustworthy user experience.
In essence, I envision a citation system where references are seamlessly integrated throughout the text, rather than clustered at the end of paragraphs. Moreover, when appropriate, links should direct users to specific excerpts that support each claim. While this approach is undoubtedly more complex and goes beyond simple summarization, it’s the standard I would expect from a search engine developed by a leader in AI technology like OpenAI.
.