Perplexity AI cites traditional news outlets like The New York Times significantly more often than other major AI search engines, according to new research analyzing millions of AI-generated answers. While most AI systems prioritize service journalism and tech reviews, Perplexity stands out for rewarding hard news and established journalism, a distinction that could reshape how publishers think about AI visibility.

LLM Pulse, a research firm tracking AI citation patterns, analyzed 5,348,765 citations from 470,380 AI answers generated between April and July 2026 across six major AI engines: ChatGPT, Google AI Mode, Google AI Overviews, Gemini, Perplexity, and Copilot. The study examined which news and media outlets these systems actually cite when answering generic questions about products, travel, and other consumer topics.

Which AI Engines Cite News Most Frequently?

The research reveals stark differences in how much each AI engine relies on journalism. Gemini and ChatGPT lead the pack, with 4.63% and 4.31% of their citations respectively pointing to news outlets. Perplexity ranks fourth at 3.10%, below Copilot's 4.24% but ahead of Google's AI surfaces. Google AI Mode and Google AI Overviews, which produce the highest volume of citations overall, lean on journalism the least, at 2.92% and 2.75% respectively.

Across all six engines combined, news and media outlets account for just 3.26% of all citations. The remaining 96.7% goes to Reddit, YouTube, retailers, product pages, academic sources, review platforms, and company websites. This suggests that when AI systems answer questions, they draw far more heavily from user-generated content and commercial sources than from professional journalism.

Why Does Perplexity Favor Traditional News Outlets?

While TechRadar dominates as the single most-cited news outlet across all engines with 14,311 citations, Perplexity's citation patterns tell a different story. The New York Times ranks 10th overall with 3,628 citations, but when Perplexity cites news sources, The New York Times appears as the third most-cited outlet at 5.7% of Perplexity's news citations, over-indexing 2.8 times compared to its overall share across all engines.

PCMag and TechRadar still lead Perplexity's citations at 10.1% and 7.1% respectively, reflecting the broader tech-first tendency of AI systems. However, Perplexity's willingness to cite traditional journalism more frequently suggests the engine may be designed with different priorities than competitors. The research notes that "if any engine rewards traditional journalism, it is Perplexity, though it still leans tech-first".

How Do AI Engines Choose Which Outlets to Cite?

The data reveals a clear pattern: AI systems overwhelmingly cite service journalism over hard news. The top-cited outlets across all engines are those built on product reviews, how-to guides, and recommendations rather than investigative reporting or breaking news. TechRadar, PCMag, Mashable, Tom's Guide, and CNET dominate the rankings, while outlets focused on breaking news and reporting, such as The New York Times, CNN, NBC News, Reuters, and USA Today, rank significantly lower.

This preference reflects the types of questions AI systems are designed to answer. When someone asks ChatGPT or Perplexity which laptop to buy or how to pick travel insurance, the system returns recommendations and explanations rather than news analysis. The outlets cited most frequently are those that specialize in exactly this kind of content.

Each AI engine also shows distinct preferences. ChatGPT sends 18.5% of its news citations to TechRadar alone, while Gemini leans hardest on PCMag at 16.5%. Google's AI surfaces favor Mashable, PCMag, and U.S. News. These preferences suggest that each engine's training data and citation algorithms have been tuned differently.

Steps to Optimize Your Content for AI Search Engines

• Maintain Data Consistency Across Platforms: AI engines cross-reference information across multiple sources to verify accuracy. If your business information, hours, or contact details differ across platforms, AI systems treat this as a trust signal and may cite you less frequently or not at all.
• Update Information in Real Time: Aggregator-based distribution systems can introduce delays of days or weeks before updates reach all platforms. Direct distribution to major publishers and niche directories ensures your information reaches AI engines quickly, improving your chances of being cited.
• Focus on Service-Oriented Content: If you operate a news or media outlet, AI systems cite service journalism, how-to guides, and product reviews far more frequently than hard news. Content that answers "how to" or "best" questions aligns better with how AI systems generate answers.

The broader implication is that AI search is reshaping the media landscape. As AI-generated answers become a primary discovery mechanism for what people read, the outlets cited in those answers gain visibility and traffic. Publishers that understand how each AI engine weights citations can optimize their content strategy accordingly.

What Does This Mean for Publishers and Businesses?

The research highlights a critical shift in how information flows to consumers. When someone uses an AI search engine instead of Google, they are no longer seeing a ranked list of websites. Instead, they receive a synthesized answer that cites multiple sources. The outlets cited in that answer become the new "front page," determining which publishers gain visibility and authority.

For businesses, the stakes are even higher. AI engines verify brand information by comparing data across hundreds of sources, including obscure directories most people have never heard of. If your business hours are correct on one platform but outdated on ten others, AI systems interpret this inconsistency as a trust problem and may decline to cite you at all.

Perplexity's higher citation of traditional journalism suggests that different AI engines may reward different types of content and sources. Publishers and businesses that understand these distinctions can tailor their strategies to maximize visibility across the AI search landscape. The question is no longer just whether you appear in Google's search results, but whether you appear in the answers AI systems generate.