
When an AI agent books your next appointment or follows up on a lead, it's easy to imagine the intelligence happening entirely inside the model. It doesn't. Most useful agents reach out to the live web, or to a business's own connected systems, to get the facts they need. And the infrastructure that makes that possible is going through a quiet but significant shakeup.
Key points
- Oxylabs, Decodo, Webshare, ScrapingBee, and Honeygain now operate together under a Vilnius-based umbrella entity called Source.
- Source reports demand across its companies has grown roughly fourfold in a year, driven by AI agents and AI search tools pulling live web data.
- The AI-driven web scraping market is projected to grow from about $10.20B in 2026 to about $23.7B by 2030, per a recent market report.
- Legal friction is escalating: Reddit's lawsuit over alleged anti-scraping evasion is proceeding against several data suppliers and AI firms.
- For small businesses, the durable play is agents grounded in your own systems and legitimate data connections rather than contested scraping pipelines.
Earlier this year, five familiar names in web scraping and proxy services, Oxylabs, Decodo (formerly Smartproxy), Webshare, ScrapingBee, and Honeygain, began operating under a single umbrella entity called Source, headquartered in Vilnius, Lithuania. According to TechRadar Pro, the move formalizes relationships that were already deeply intertwined. Honeygain became Oxylabs' main residential proxy sourcing partner back in 2021, Oxylabs acquired Webshare in 2022, and it picked up ScrapingBee in June 2025. What looked like a competitive market was, in several cases, a family of companies that had been marketed and reviewed as direct rivals. TechRadar notes that three of its own nine best-proxy picks now share ownership. Source positions itself as a shared-services backbone, handling patents, procurement, hiring, and compliance, while each brand keeps its own team, product roadmap, and customers. Whether they keep competing on price, and whether common ownership gets disclosed clearly to buyers, remain open questions.
The more interesting number is the demand curve. Source claims requests across its companies have grown roughly fourfold in a year, and its infrastructure reportedly handles about 1.5 million requests per second. The company attributes the surge to AI agents, next-generation search tools, and frontier AI products that need continuous access to live web data. A recent market report puts the AI-driven web scraping market at roughly $8.24 billion in 2025, growing to about $10.20 billion in 2026, with projections reaching about $23.7 billion by 2030 at a compound annual growth rate near 23.5 percent. Those figures come from the cited sources and haven't been independently verified, but the direction is clear. Agents that answer phones, qualify leads, and run workflows need current information, and someone has to supply it at scale. That supply chain is now big business.
Consolidation isn't only about growth. It's also about legal exposure, and this sector has plenty. In 2021, a Texas jury awarded Bright Data roughly $7.5 million in a patent case against Oxylabs. In October 2025, Reddit sued Oxylabs, Perplexity, SerpApi, and AWMProxy over alleged evasion of its anti-scraping measures, and in July 2026 a federal judge allowed Reddit's core claims against SerpApi and Perplexity to proceed, with Oxylabs still a listed defendant as of spring 2026. Source has argued publicly that per-deal licensing regimes would shut out all but the largest players. That's a self-interested position, but it's also a substantive one, because it sits at the center of an unresolved debate: who gets access to public web data, on what terms, and whether the answer leaves room for anyone outside the biggest AI companies. The outcome of the Reddit litigation could shape that answer for years.
If you run a trade business, a law practice, a clinic, an insurance office, or an ecommerce brand, you might reasonably ask what any of this has to do with you. Quite a bit, actually. The same market report that forecasts the scraping boom specifically cites rising SME interest and no-code tools as growth drivers. Small teams want working agents without building them in-house, and they want outcomes: calls answered, appointments booked, leads followed up, busywork automated. But the consolidation story carries a practical lesson for anyone deploying agents. If your agent's usefulness depends on contested scraping pipelines, its reliability and cost are exposed to lawsuits and licensing fights you have no control over. The more durable approach is agents grounded in your own legitimate data sources: your CRM, your calendar, your phone system, your inbox. Agents built on data you already own and integrations you're entitled to use don't inherit someone else's legal risk. That's the philosophy behind our own done-for-you builds at Agents by AIQ. We connect agents to the tools a business already runs, month to month, and the client owns everything at the end.
There's a useful mental model here. AI agents are the storefront; the web data layer is the warehouse and the freight lines behind it. When five freight companies merge and report fourfold demand growth, that tells you the storefronts are doing real business. It also tells you the warehouse is where bottlenecks, price shifts, and legal disputes will land first. If you're evaluating agents for your business, ask where the data comes from. A vendor who can't answer that clearly, or whose answer leans on gray-area scraping, is asking you to carry risk you didn't sign up for. A vendor whose agents read your own systems through proper integrations is building on ground you control.
The formation of Source, the fourfold demand growth its companies report, and the ongoing litigation over web data access all point to the same conclusion: AI agents have moved from novelty to infrastructure, and the infrastructure has a supply chain with real consolidation and real legal risk. For small and mid-size businesses, the smart move isn't to become a data plumbing expert. It's to work with builders who ground your agents in your own systems and legitimate connections, so your answered calls and booked appointments don't depend on the outcome of a courtroom fight in California. If you'd like to scope what that could look like in your business, book a call and we'll sketch it out with you.