Perplexity Raises $500M at $14B to Push Agentic Search
Perplexity AI has closed a $500 million Series D led by SoftBank at a $14 billion valuation. The funds are earmarked for expanding its agentic search platform, international markets, and a new enterprise tier expected later this summer.
Original sourcePerplexity AI announced the close of a $500 million Series D funding round led by SoftBank, valuing the company at $14 billion — more than double its $6.5 billion valuation from its last raise in late 2024. The round positions Perplexity as one of the most heavily capitalized AI search companies outside of the major platform incumbents, and signals continued investor appetite for search alternatives built on top of large language models.
The company says the capital will accelerate three initiatives: expanding its agentic search capabilities, growing into international markets where English-language search dominance is less entrenched, and launching a dedicated enterprise tier later this summer. The agentic angle is the most strategically significant — Perplexity has been quietly shipping features that allow the platform to take multi-step actions on a user's behalf, moving beyond answer retrieval toward task execution.
The $14 billion valuation is a notable milestone, but it arrives alongside real competitive pressure. Google has significantly improved AI Overviews in Search, OpenAI's ChatGPT search handles a substantial and growing query volume, and Microsoft continues to push Copilot into enterprise workflows. Perplexity's bet is that a purpose-built search-and-action product can outmaneuver features bolted onto legacy platforms — a plausible but unproven thesis at this scale.
The enterprise tier is the near-term business story to watch. Perplexity has grown its consumer and prosumer user base meaningfully, but enterprise contracts are where $14 billion valuations need to find their revenue justification. The product details of that tier — pricing, admin controls, data handling, and API surface — will matter more than the funding headline in determining whether this round was prescient or premature.
Panel Takes
The Founder
Business & Market
“The buyer question is the whole game here: Perplexity's consumer growth is real, but consumer love doesn't justify a $14B valuation — enterprise contracts do. The summer enterprise tier launch is do-or-die for the unit economics story, and the details that matter aren't in this press release: what's the ACV, who owns the procurement motion, and does it come out of IT budget or departmental AI spend? The moat concern is genuine — if the product is fundamentally a better search interface, Google shipping marginally better AI Overviews is an existential threat, not a competitive nuisance.”
The Skeptic
Reality Check
“A $14B valuation for a company whose core product — answer synthesis over web results — is a feature that Google, OpenAI, and Microsoft are all shipping natively is a bet that execution beats incumbency, and that bet has a poor historical win rate in search specifically. The agentic framing is the right strategic pivot, but 'agentic search' is doing a lot of work here without a concrete definition of what agents Perplexity runs, at what reliability, and on what task types that competitors can't already handle. What kills this in 18 months: OpenAI tightens the ChatGPT search integration with actions, and Perplexity is left holding a very expensive consumer app.”
The Futurist
Big Picture
“The thesis Perplexity is betting on is falsifiable: that query-as-task-initiation will replace query-as-information-retrieval as the dominant search behavior within three years, and that this transition favors a purpose-built product over search platforms retrofitting agent layers onto link indexes. That's a real and plausible bet — the trend line is user tolerance for multi-step delegation to AI, which is measurably increasing. The dependency that has to hold: LLM tool-use reliability needs to cross a threshold where failed agent actions don't create more friction than just Googling it, and we're not there yet at scale for complex tasks.”
The PM
Product Strategy
“The job-to-be-done for Perplexity's consumer product is clear and well-executed — 'get me a sourced answer without clicking ten links' — but the agentic expansion introduces a second job that doesn't obviously share the same user or the same workflow. The enterprise tier announcement with no shipping date, no feature list, and no pricing is a roadmap slide dressed up as product news; I'll evaluate it when it exists. The product risk isn't ambition, it's that 'agentic search' as a job-to-be-done is still undefined enough that Perplexity might ship a complete product for a job users haven't fully committed to hiring anyone to do yet.”