If you have ever winced at your monthly AI bill, here’s some good news. According to a report by The Wall Street Journal, OpenAI is considering drastically lowering the prices it charges users as it fights to win customers from its rival, Anthropic.

The company is weighing significant cuts to its token pricing, the unit AI firms use to bill for their products. Interestingly, the move is in anticipation of similar cuts OpenAI expects from Anthropic. So whichever AI service you use, your bills should get smaller.

Why is OpenAI suddenly feeling generous?

The answer is simple: businesses are tired of paying sky-high prices for AI. Heck, there have even been reports of AI costing companies more than actual employees. Even OpenAI CEO Sam Altman admitted at a recent event that costs had become “a huge issue,” adding, “I think we’ll have a lot of ways we can help people get more value for less spend.”

But that’s not all. OpenAI is also facing stiff competition. Anthropic’s revenue surged after its coding tool, Claude Code, went viral among software engineers, and the five-year-old startup surpassed OpenAI’s valuation for the first time. OpenAI has since made its own coding tool, Codex, a company focus, but it’s still far behind the competition.

Some corporations poured so much money into AI coding tools that their leaders are now reining in spending. An Uber executive said the company had already maxed out its 2026 budget for agentic AI. These comments have sparked a Silicon Valley debate about tokenmaxxing, the practice of burning through as many tokens as possible to boost productivity, even when it doesn’t generate returns.

Then there’s Google. Its Gemini models, especially the budget Flash tiers, undercut both ChatGPT and Claude on price, and its business plans cost nearly half of what OpenAI charges, adding even more competitive pressure.

What does a price war mean for you?

For the companies, it’s risky. Both companies already lose billions on computing costs, and both have confidentially filed for IPOs. Slashing prices right before facing public investors will be the first real test of their business models.

For users, it’s good news. They will soon see a drastic reduction in their AI costs. Competition is always good for consumers, and a price cut is one of the big benefits. So sit back and let the AI giants fight it out, because for once, we are the ones who win.

TL;DR

Perplexity AI announced a platform at Computex that dynamically routes AI inference between PCs and cloud servers in real time, acting as an “air-traffic controller” for AI tasks. The chip-agnostic system targets the cost crisis of centralised inference as Perplexity’s revenue hits $500 million.

Perplexity AI has developed a platform that dynamically splits AI workloads between personal computers and cloud servers, deciding in real time which tasks can run locally on a PC’s processor and which need the power of data centre hardware. CEO Aravind Srinivas announced the system at Computex in Taipei on Tuesday, describing it as an “air-traffic controller for AI tasks” designed to reduce the cost of inference, the process of running trained AI models to generate responses.

“You don’t want all your compute centralised in servers and everything running through the largest models,” Srinivas said in a Bloomberg Television interview. “You’re already reading reports of how people are freaking out about their cost. Some people are spending half a billion dollars per month. What you actually want is efficient value per watt per user.”

How it works

The system evaluates each AI task and routes it to the most efficient compute layer. Simple operations that modern PC processors can handle, such as summarisation, formatting, or lightweight classification, run locally without touching the cloud. More complex tasks that require large model inference, such as multi-step reasoning or retrieval-augmented generation across large datasets, get routed to cloud servers. The routing decision happens in real time, invisible to the user.

The practical effect is that Perplexity can serve more users at lower cost by offloading a portion of inference work to the billions of PCs already in circulation. As AI inference demand strains data centre capacity and drives utilities to plan $1.4 trillion in grid upgrades, distributing compute to the edge is both an economic and infrastructure necessity.

Srinivas made the announcement alongside Intel CEO Lip-Bu Tan, whose company leads the market for PC processors and has a commercial interest in making PCs a meaningful AI compute layer. However, Srinivas said the platform is “chip agnostic” and works with Nvidia processors as well. Nvidia highlighted the same edge-inference trend at Computex with its new RTX Spark platform for AI-powered laptops and desktops.

The cost problem

Srinivas’s reference to companies “spending half a billion dollars per month” on AI compute is not hyperbole. OpenAI’s infrastructure costs have been widely reported at that scale, and Anthropic’s projected $10.9 billion in Q2 revenue comes with substantial compute expenses that compress margins. The energy and cost burden of centralised AI inference is one of the defining constraints of the current AI boom.

Perplexity’s approach inverts the assumption that AI inference must happen in the cloud. By treating the PC as a first-class compute node rather than a thin client, the company can reduce its own server costs while potentially delivering faster responses for tasks that run locally. The tradeoff is complexity: the routing system must accurately assess task difficulty in milliseconds, and the quality of local inference depends on the user’s hardware capabilities.

Revenue efficiency

Perplexity’s financial trajectory underscores why cost efficiency matters. Srinivas posted on X in April that the company’s revenue grew fivefold, from $100 million to $500 million, while headcount increased just 34%. That ratio, roughly 15x revenue growth per employee added, reflects both the leverage of AI-native business models and Perplexity’s position as an aggregator that routes queries across multiple AI providers rather than training its own frontier models.

“Every time any of the AI gets better, our unified system also gets better because we route across all of them,” Srinivas said. The AI-native growth rates that are drawing capital away from traditional SaaS companies are partly enabled by this kind of architectural efficiency, where the product improves as its underlying providers improve, without proportional cost increases.

The hybrid compute platform extends that logic to hardware. If Perplexity can use the compute already sitting on users’ desks to handle a meaningful share of inference work, it reduces marginal cost per query and improves response latency for lightweight tasks. As AI moves deeper into enterprise workflows, the economics of who pays for the compute, the cloud provider, the AI company, or the user’s own hardware, will become a critical competitive variable.