Five of the web's biggest data-scraping brands now sit under one $3.6 billion umbrella, and AI agents are the reason why

Source is a Lithuanian umbrella organization for five web-data companies: Oxylabs, Decodo, Webshare, ScrapingBee, and Honeygain. It was created as demand surged for services that help software access public web data. The unusual part is that these brands still present themselves as separate businesses.

Source says it has no customers or products of its own. Instead, it manages patents, procurement, hiring, and compliance. Oxylabs serves enterprises and AI labs, while Decodo targets mid-market customers. ScrapingBee focuses on developer APIs, Webshare serves professional consumers and solo builders, and Honeygain sources residential bandwidth from ordinary users.

The arrangement matters because several brands were previously reviewed as competitors. TechRadar's proxy guide includes Decodo, Oxylabs, and Webshare as separate picks. Source says each keeps its own strategy, team, customers, and market presence, but common oversight now links them formally.

Source claims demand for services across its organizations has risen fourfold in a year. The growth is tied to AI agents, next-generation search tools, and frontier AI products that need current information from the web. This makes web access a rapidly expanding infrastructure market rather than only a traditional scraping service.

The group says its systems now process 1.5 million requests per second. A request is an individual call for information or a web resource, so the figure indicates very high operational scale. It also says the companies are expanding into web search APIs designed specifically for AI agents.

The numbers show both opportunity and pressure. More automated software is requesting fresh data, increasing demand for reliable access. At the same time, the scale raises questions about responsible use, website defenses, licensing, and whether a large supplier group could gain too much influence over web access.

AI agents are software systems that can carry out multi-step tasks, such as searching, comparing, and retrieving information, rather than waiting for a person to perform every action. In this article's context, they use web services to obtain current information for search and AI applications. Their value depends partly on reaching data that changes after a model was trained.

For example, an agent might send a search request, receive results, and then ask for pages to be retrieved or analyzed. Oxylabs already offers a Web API that combines a search endpoint with a scraping endpoint for agents. Source says it plans to expand dedicated AI Search APIs, building on that existing work.

Continual access matters because websites, prices, news, and public records can change. The article does not specify particular agent tasks, so those examples are general context. It does show that AI products are pulling live web data at scale, creating fourfold demand growth for Source's services.

The brands are divided by audience. Oxylabs serves enterprises and AI labs. Decodo, formerly Smartproxy, targets the mid-market. ScrapingBee sells developer APIs, Webshare focuses on professional consumers and solo builders, and Honeygain pays ordinary people for spare bandwidth used in residential proxy sourcing.

Their different customer segments make separate products possible. Source handles shared functions such as patents, procurement, hiring, and compliance, while each brand keeps its own team, product strategy, and customer base. That structure lets the companies maintain distinct offerings and market identities even under one umbrella.

However, the separation does not erase overlap. TechRadar has reviewed Decodo, Oxylabs, and Webshare as separate proxy picks, and Oxylabs was described as one of Decodo's toughest competitors. Source says they remain independent in strategy and market presence, but customers now know common oversight connects brands previously presented as alternatives.

For customers, the immediate issue is transparency. They may still buy from separate brands, but those brands now belong to one formally connected organization. Customers will want to know whether pricing, policies, data practices, and support truly remain independent. The article specifically identifies competition on price and disclosure as questions to watch.

Source could also combine expertise and purchasing power while preserving specialized products. Its stated role covers patents, procurement, hiring, and compliance. The group says it offers a unified voice for responsible access to public web data. That could help coordinate industry positions, but it may also make the umbrella more influential.

Legal disputes give the arrangement added weight. Reddit sued Oxylabs with other companies over alleged evasion of anti-scraping measures. A separate patent case brought Bright Data nearly $7.5 million from Oxylabs. Source's shared oversight could affect how the group responds to such scrutiny, although the article does not predict legal outcomes.

Oxylabs acquired Webshare in 2022 and ScrapingBee in 2025. Honeygain became Oxylabs' main residential proxy sourcing partner in 2021. These deals and partnerships created practical links among several brands before the formal umbrella appeared. Decodo's connection was less clear until Source's launch officially placed it under the same organization.

Leadership changes reinforced that relationship. Source CEO Vytautas Savickas ran Decodo until March 2026, then moved to lead the Oxylabs group. Decodo founder Julius Černiauskas became chairman. The article describes the result as a close interlinking and a marriage of convenience rather than a complete merger.

The legal background includes Reddit's October 2025 lawsuit against Oxylabs and others, alleging they bypassed Reddit's anti-scraping measures through Google results. Oxylabs denied the allegations and said it would defend itself. Earlier, a Texas jury awarded rival Bright Data nearly $7.5 million in a patent case against Oxylabs.

A proxy server is an intermediary that sends a software request to a website and returns the website's response to the requesting application. Web scraping is the automated collection of information from web pages. Used together, they give software a way to request and organize web content without a person manually visiting every page.

A scraping system can send a request through a proxy, receive a page or search result, and extract the needed information. The article describes services offering search and scraping endpoints, including Oxylabs' Web API for agents. Honeygain supplies residential proxy sourcing by paying ordinary people for spare bandwidth.

This infrastructure helps applications retrieve changing public data, including systems described as AI agents. The article reports 1.5 million requests per second across Source's systems, showing the scale involved. It also highlights disputes over anti-scraping measures, patents, and responsible access, so technical ability does not settle whether every use is permitted.