In Battle of AI Search Engines, Perplexity Has Its Sights Set on One Main Rival

Google Gemini is the "main rival" that Perplexity is focused on competing against in the rapidly evolving AI search landscape, according to company CEO Aravind Srinivas.

The strategic positioning marks a significant shift in how AI search startups view the competitive field, End of Miles reports, as Perplexity explicitly names its primary target among the tech giants scrambling for dominance in AI-powered search.

The battle for search supremacy

During a recent Reddit AMA (Ask Me Anything), Srinivas directly addressed how Perplexity plans to compete with the "deep research" capabilities being offered natively by large language model providers like Google and OpenAI.

"This is a good question. Let me be honest in admitting that Deep Research removes the need for having a really fast search index + ranking infrastructure which is something we heavily invested in for making the core default Perplexity product better than our competitors." Aravind Srinivas, Perplexity CEO

The Perplexity founder acknowledged that as major tech companies introduce their own AI search features, the differentiation his company initially built around its search index faces new challenges. Yet rather than seeing all AI providers as equal competitors, he specifically highlighted Google's AI offering as the one to beat.

"I primarily see Google Gemini as the main rival here, and I think the way we will make it better is to just get the finer details right: better tool use capabilities, proprietary data providers, and cheaper costs by making use of open source models." Srinivas

The open source advantage

While Google and other tech giants invest billions in developing proprietary AI models, Perplexity is taking a different approach. The startup is betting on enhancing open source models rather than building its own from scratch – a strategy Srinivas believes will provide cost efficiencies and flexibility.

This reflects a broader trend in AI development, where companies are finding ways to compete not just on raw model capabilities but on specialization, cost-effectiveness, and user experience.

"In my 2.5 years of running Perplexity, one thing I have realized is no one has a long-term model advantage. The product is the ultimate place to compete on. And open source models always keep catching up or improving on the closed ones, with better cost-effectiveness and speed and customization." Srinivas

The differentiation strategy

While acknowledging Google's formidable position, the AI entrepreneur outlined a three-pronged approach to differentiation: superior tool use capabilities, proprietary data sources, and cost efficiency through open source model optimization.

The company is already putting this strategy into practice. Srinivas revealed that Perplexity is currently running an A/B test with a post-trained version of DeepSeek-v3 that "seems to be better than GPT 4o" – OpenAI's latest model.

"We are not trying to have the best model at any point; but rather the ability to post-train and improve upon anything that exists out there within a few months." Srinivas

This focus on rapid adaptation rather than building foundational models from scratch represents a strategic calculation that the future of AI search will be determined not by who has the most sophisticated core model, but by who can most effectively customize and deploy existing technologies.

Beyond search to assistants

Looking ahead, the Perplexity chief believes user behavior will shift from primarily asking AI systems questions to having AI perform tasks autonomously.

"I think if the progress in AI reasoning models continues, search itself might be more for the curious minded. And people are going to be using AIs largely for workflows and tasks which will require plenty of search in the loop, but will move to the abstraction of asking AIs to do work for you, instead." Srinivas

The company's upcoming Comet browser and focus on "browser agent" technology represent steps toward this vision – potentially carving out competitive advantages that even tech giants with massive resources might struggle to match quickly.