4 out of my last 6 Perplexity searches were misleading or false.
Unlike the other stats in my last search, the numbers above are accurate.
My favorite “glitch” of modern search – also visible in ChatGPT and AI overviews in Google – is:
– interpreting or quoting data that doesn’t exist,
– mixing numbers across different topics, or…
– quoting articles that aren’t proofread.
As AI keeps scraping search without proper recognition on pay, we end up with:
> paid reddit actors producing fake content and recommendations
> programmatic SEO randomly producing random articles
> random data seen somewhere that isn’t validated
If you look up networking stats for B2B, the first 4 results, even while seeming “authoritative” with links to posts, would be a mix of AI generated content that was indexed and random numbers not quoting sources.
One of these post links to a LinkedIn study… but the quoted metric doesn’t exist there.
—
Really not bashing AI on purpose, but what did anyone expect scraping the Internet and running an ongoing crawler now, whereas users intentionally pollute public data as revenge?
