David Bridges
Perplexity has recently launched an innovative Deep Research tool, leveraging a customized version of DeepSeek R1 to enhance its functionalities. This powerful tool conducts comprehensive research and analysis on behalf of users, scouring the vastness of the internet to compile detailed reports on various topics. The introduction of this feature positions Perplexity alongside industry giants such as Google and OpenAI, both of which have developed their own research tools named Deep Research for their platforms, Gemini and ChatGPT. Meanwhile, xAI has opted for a different naming strategy with its research tool, calling it Deepsearch, showcasing the competitive landscape of AI-driven research solutions.
While established players like Google and OpenAI depend on their proprietary models, Perplexity distinguishes itself by utilizing a modified version of the open-source DeepSeek R1. Aravind Srinivas, the CEO of Perplexity, highlighted the cost-effectiveness of this approach, stating that they can provide something akin to Deep Research at a significantly lower price point—between 10 to 100 times cheaper—thanks to their customized version of R1. Following this announcement, Srinivas confirmed the successful implementation of the tool with a simple yet effective affirmation on social media, indicating their commitment to making research more accessible.
Mashable Light Speed
Perplexity is making its Deep Research tool available free of charge, albeit with certain limitations. Users who opt for the free version will have access to a limited number of answers per day, while those who subscribe to the Pro plan at $20 a month will enjoy unlimited access to this robust research feature. This approach sets Perplexity apart from competitors like Google Gemini, ChatGPT, and Grok 3, which exclusively offer their research tools to paid subscribers, thereby enhancing Perplexity’s appeal among users looking for cost-effective solutions.
Perplexity’s adaptation of the DeepSeek R1 model allows for significant customization, as the open-source nature of R1 enables developers to modify it to suit their needs. Recently, Perplexity introduced its own open-source variant known as R1 1776, which has been “post-trained” to deliver uncensored, unbiased, and factual information. This move directly addresses concerns regarding censorship, particularly those alleging that the original R1 model suppresses critical perspectives on the Chinese government, thereby enhancing the credibility and transparency of their research outputs.
Despite its potential, the Deep Research tool from Perplexity is not without its challenges. Recent assessments by Decoder revealed inaccuracies, such as misattributing the term “stochastic parrots” to AI researcher Gary Marcus, when it was actually coined by Emily M. Bender in her research. Additionally, users have reported that the tool occasionally provides outdated or inaccurate data, raising concerns, especially since Perplexity promotes it as a resource for investment and market analysis. Acknowledging these issues, Srinivas emphasized the importance of data accuracy in finance, declaring that the stakes are indeed high for users relying on accurate information.
This situation underscores a persistent challenge faced by all large language models (LLMs)—the phenomenon known as hallucination. This issue remains a significant obstacle in the field of AI, and users are advised to exercise caution when utilizing these tools, as inaccuracies can lead to misguided conclusions or decisions.
Topics
Artificial Intelligence
DeepSeek
var facebookPixelLoaded = false;
window.addEventListener(‘load’, function(){
document.addEventListener(‘scroll’, facebookPixelScript);
document.addEventListener(‘mousemove’, facebookPixelScript);
})
function facebookPixelScript() {
if (!facebookPixelLoaded) {
facebookPixelLoaded = true;
document.removeEventListener(‘scroll’, facebookPixelScript);
document.removeEventListener(‘mousemove’, facebookPixelScript);
!function(f,b,e,v,n,t,s){if(f.fbq)return;n=f.fbq=function(){n.callMethod?
n.callMethod.apply(n,arguments):n.queue.push(arguments)};if(!f._fbq)f._fbq=n;
n.push=n;n.loaded=!0;n.version=’2.0′;n.queue=[];t=b.createElement(e);t.async=!0;
t.src=v;s=b.getElementsByTagName(e)[0];s.parentNode.insertBefore(t,s)}(window,
document,’script’,’//connect.facebook.net/en_US/fbevents.js’);
fbq(‘init’, ‘1453039084979896’);
fbq(‘track’, “PageView”);
}
}
Daniel Mercer
