A new model for AI advertising emerges as agents could replace human attention

Perplexity AI's founder Aravind Srinivas has outlined a radical transformation in digital advertising, where artificial intelligence agents, rather than humans, could become the primary target for advertisements. This marks a significant shift from traditional digital advertising models that have dominated the internet for decades.

According to Srinivas, in an interview conducted on December 30, 2024, the future of advertising could involve AI agents acting as intermediaries between brands and consumers. "The user never sees an ad," Srinivas explained. "The different merchants are not competing for users' attention; they're competing for the agents' attention."

The proposed system would function differently from current digital advertising platforms. When a user makes a request - for example, planning a two-night stay at a hotel - multiple travel companies, platforms, and airlines would compete for the AI agent's attention in the backend, without the human user ever seeing traditional advertisements.

This transformation builds on Perplexity AI's recent growth trajectory. The company has seen its annual recurring revenue increase from $3-4 million at the beginning of 2024 to approximately $70 million by year-end. Daily queries have grown from 2.5 million to 20 million during the same period.

The technical implementation would involve significant computational resources. According to Srinivas, AI agents could "spend a lot more compute power" analyzing offers and making decisions based on user preferences. Users could instruct their agents to ignore certain brands, with these preferences remaining confidential between the user and their agent.

The model presents multiple potential revenue streams. Srinivas identified four distinct monetization levels: direct subscriptions for AI assistant services, task-based usage fees, transaction commission fees, and a new form of advertising revenue where brands bid for agent attention.

Industry responses to this vision have been mixed. Some technology experts have raised concerns about potential conflicts of interest and the need for transparency in agent-based decision making. Others have questioned whether consumers would be willing to delegate significant purchasing decisions to AI systems.

The proposed advertising model aligns with broader industry trends toward AI-driven automation but represents a fundamental shift in how digital advertising could function. Currently, major platforms like Google generate significant revenue from showing ads directly to users. The new model would potentially operate with lower profit margins than traditional search advertising, which Srinivas acknowledges as "pretty much impossible" to match.

Perplexity AI, valued at $9 billion, continues to position itself as an innovator in the artificial intelligence sector. This advertising model represents part of the company's broader vision for AI agents that can take actions on behalf of users, moving beyond simple question-answering to more complex tasks and decisions.

The technical challenges of implementing such a system remain substantial. The company's focus for 2025 will be on improving reliability and expanding the capabilities of AI agents to handle increasingly complex tasks and decisions.

For the advertising industry, this proposed transformation raises fundamental questions about the future of digital marketing and the role of human attention in advertising models. The implementation timeline and broader market adoption of such systems remain uncertain, as both technical capabilities and consumer trust will need to develop significantly.

As the artificial intelligence sector continues to evolve, Perplexity AI's vision for advertising represents one potential future direction for the industry, though its realization will depend on numerous technological, economic, and social factors.

Perplexity AI emerges as next-generation search engine

Perplexity AI, founded in August 2022, has rapidly emerged as a significant player in artificial intelligence search, reaching a $9 billion valuation in December 2024. The San Francisco-based company represents a fundamental shift in how people access and process online information.

At its core, Perplexity AI functions as a conversational search engine powered by large language models (LLMs). According to company documentation, the platform synthesizes information from across the web and provides answers with inline citations, distinguishing itself from traditional search engines that primarily return lists of links.

The founding team brings substantial expertise from leading technology companies. CEO Aravind Srinivas previously worked as an AI researcher at OpenAI, while Chief Technology Officer Denis Yarats served as an AI research scientist at Meta. Andy Konwinski, another co-founder, was part of the founding team at Databricks, and Chief Strategy Officer Johnny Ho brought experience from Quora and Wall Street trading.

According to Bloomberg News, Perplexity's growth has been remarkable, with annual recurring revenue increasing from $3-4 million at the beginning of 2024 to approximately $70 million by year-end. Daily queries have grown from 2.5 million to 20 million during the same period.

The platform operates on a freemium model. The free version utilizes the company's standalone LLM based on GPT-3.5 with web browsing capabilities. According to company materials, the Pro subscription provides access to more advanced models including GPT-4, Claude 3.5, Grok-2, and Llama 3, along with features like file upload and analysis capabilities.

In October 2024, Perplexity expanded its offerings by introducing Spaces, an AI-powered collaboration hub allowing users to create customized knowledge spaces combining web searches with personal file integration. According to technical documentation, users can upload up to 50 different documents, with a 25MB size limit per file.

The company has attracted significant investment from prominent figures in technology. According to regulatory filings, notable investors include Jeff Bezos, Nvidia, Databricks, Bessemer Venture Partners, and technology leaders such as Susan Wojcicki, Jeff Dean, Yann LeCun, and Andrej Karpathy.

Recent developments include the November 2024 launch of a shopping hub, backed by Amazon and Nvidia, which provides product recommendations in response to user queries. The company has also introduced finance-related features, including real-time stock quotes and company earnings data, sourcing financial information from Financial Modeling Prep.

However, Perplexity has faced challenges regarding content usage. According to court documents, in October 2024, The New York Times sent a cease-and-desist notice, claiming copyright infringement through unauthorized use of its content. Dow Jones and the New York Post have filed similar lawsuits, raising questions about content attribution and usage rights in AI-powered search.

The platform's approach to web crawling has also drawn scrutiny. According to investigations by Wired magazine and web developer Robb Knight in June 2024, questions arose about the company's compliance with the robots.txt standard, which allows websites to control web crawler access. In response, Srinivas stated that Perplexity relies on both proprietary and third-party web crawlers, explaining the complexity of content access mechanisms.

Looking ahead, Perplexity plans to introduce advertising on its search platform. According to public statements by company leadership, this move aims to create new revenue streams while maintaining user experience quality. The company has also launched a publishers' program to share advertising revenue with content partners, addressing concerns about fair compensation for content usage.

The rapid rise of Perplexity AI reflects broader changes in information access and processing. As traditional search engines face new competition, the company's growth suggests increasing user demand for AI-powered tools that can synthesize and contextualize information from multiple sources. The platform's success also raises important questions about content rights, fair compensation for publishers, and the future of web search technology.

Who is Aravind Srinivas?

Aravind Srinivas, the 30-year-old co-founder and CEO of Perplexity AI, represents a new generation of AI entrepreneurs reshaping the technology landscape. According to his academic record, Srinivas completed his education at the Indian Institute of Technology Madras before pursuing a Ph.D. in Computer Science at the University of California, Berkeley.

According to research publications, Srinivas has made significant contributions to the field of artificial intelligence. His academic work, cited over 10,000 times according to Google Scholar, spans deep learning, reinforcement learning, and contrastive learning. His most influential paper on Decision Transformers, published in 2021, has received more than 1,700 citations.

Before founding Perplexity AI, Srinivas accumulated experience at several leading AI research institutions. According to his professional history, he worked as a research scientist at OpenAI, completed research internships at DeepMind and Google, and contributed to breakthrough projects in computer vision and machine learning.

According to TIME magazine's profile in September 2024, despite Perplexity's rapid success, Srinivas maintains a notably understated demeanor that contrasts with the typical Silicon Valley CEO persona. The magazine noted his preference for focusing on technical achievements rather than personal publicity.

Srinivas has emerged as a prominent voice in the ongoing debate about skilled immigration in the United States. According to recent statements, he has advocated for merit-based immigration while acknowledging his own status as an immigrant entrepreneur on a visa. His company, Perplexity AI, has created over 100 jobs in the United States, according to public statements from company executives.

Under his leadership, Perplexity AI has grown from a startup to a company valued at $9 billion in just over two years. According to Bloomberg News, the company's growth has been remarkable, with its annual recurring revenue increasing from $3-4 million to approximately $70 million during 2024.

The entrepreneur has also taken public positions on responsible AI development. According to company materials, Srinivas has implemented policies for transparent source attribution in Perplexity's AI responses and initiated a revenue-sharing program with publishers whose content the platform utilizes.

As an angel investor, Srinivas has backed various AI startups, including companies working on AI voices, foundation models, and software engineering tools. According to his professional profile, his investment portfolio spans diverse areas of artificial intelligence technology, from multimodal AI to AI-powered fitness coaching.