Perplexity is a search engine powered by AI which aims to provide precise and relevant results by citing sources and examples in its summaries. It is a powerful tool that shows how we could reinvent search engines using AI and leave Google spam results Behind for good.
The up -to -date results of Perplexity, which are combined with summaries and quotes, give the impression that the perfect replacement for Google. I use Google Daily for work, so I tried to replace it with perplexity for a week to see if it could become a permanent addition to my life.

Related
Quotes, relevant questions and summaries make research a breeze
During the last week, I completely replaced Google Search by perplexity. Whether I was looking for advice for my next hobby project or I was heading for my next article, I rather connected my request to the PERPLexity application or browser.
Perplexity shone in my personal life when I didn’t know exactly what I was looking for. For example, after being impressed by a new beer, I wanted to discover the brewery. However, I forgot the name of the brewery, and all I knew was that he was based in Nottingham and the beer that I had a relatively high alcohol content.
Connect these waves in perplexity rewarded me with the brasserie that I was looking for in the first result. Google managed to list them in second position but piled up my screen with information I didn’t need at the time. It was not a unique situation, perplexity managed to provide me with responses without bloating to my requests which were generally as precise as those of Google.


One of the most notable perplexity characteristics is its quotation characteristic. After each segment, Perplexity adds a citation link to its source. This is useful for personal use (the quote from the brewery took me directly to their website), but crucial for professional work. I tend not to use generative AI tools in my research because of its trend to hallucination, but quotes from perplexities allow me to quickly check its declarations.
So far, so good. The perplexity had impressed me in one day by his relevant answers and without bloating. His related requests helped me dive deeper into my subject, and the quotes helped soothe my concerns to use a search engine fueled in AI. But he quickly turned into the sour sense like the Common problems with generator AI Raised the head after only a day or two.
The bad: I had to verify the perplexity of the facts and its source
Twice as much research for the same result
During my research for an article, I wanted to reveal that I had not missed any key fact on the next Google I / S Developer conference. I entered “Gemini at Google I / O 2025” and a quick analysis of the results confirmed what I had already written. But I obviously cannot count on a summary of the AI, so I spend the next 15 minutes to check its sources.
The first Declaration of Perplexity cited two websites, the support plan and the technological radar. Although I confirm that he had represented the article of Radar Tech, I was wary of the lack of paternity and date of publication. These suspicions were confirmed when I checked the article Radar Tech. The assistance plan had torn off its article directly from Tech Radar, which was listed as the second source of Perplexity’s summary. The websites tearing articles of larger publications without attributing the original author are nothing new and easy to spot for humans, but Perplexity had read it as a legitimate source and prioritized it on the original.
This problem has emerged a few times during my week. Perplexity quoted sources that plagued items, often from websites that the search engine had previously mentioned. It immediately led me to question all my results. Perplexity did not provide me with a balanced summary because it scratches data from the same text several times.
Associated with this problem came from good old -fashioned information. The same request returned a chip claiming that Google planned to introduce AI previews alongside its research results. It struck me as strange, because I knew it had already happened (and was later retracted). Indeed, the article cited was from May 2024, but Perplexity said that the functionality was still underway.
Everything above meant that I could no longer trust perplexity for specific summaries. I had to verify each statement thoroughly, which ended up taking me more time than googling the question and writing a summary myself. These summaries gave me an overview which was useful for a little more than a starting point.
Personal time research has also been filled with inaccuracies. I spotted an absurd quantity of errors in this search for Warhammer models to come, because the results were scratched from badly written and inaccurate fans. Why did perplexity not use the information of the official site? No idea. Interestingly, the preview of Google AI provided more vague results, but more precise results. However, he has still failed to scratch the original source.


The Use: Copyright prosecution and violations
Perplexity does not respect its sources
At the end of my week perplexity, I was ready to abandon it. Although it is a fantastic research assistant when he worked, he suffered from the same problems as all other AI tools. Unless the problems around LLM are soon resolved, I cannot count on him.
Even if a miracle occurs and the perplexity ceases to citing plagiarized work and unrelevant items, I will still not use it. The surveys reported that perplexity is Ignore the robot exclusions protocol (which allows websites to stop web robots) using third -party web robots. Perplexity is also subject to claimed proceedings Copyright and brands violation.
Will I continue to use perplexity? No
Perplexity offers clever features which, I think, are a real improvement in the search engine formula. His past research library is much more useful than Google’s research history, and the function of relevant questions helped me think of new questions. Unfortunately, I could not test his reverse image search tool (which forced me to register for a pro subscription of $ 20 / month).
The problem is that I cannot count on its summaries, and its plagiarism of content raises serious ethical concerns. If perplexity is something, AI is not even about to reinvent the search engine.