LXT Challenges Crowd-Sourcing Giants with New 'Control-First' Data Service

TORONTO, ON – July 16, 2026 – In a significant move aimed at reshaping how artificial intelligence is trained, AI data provider LXT today launched its Crowd-as-a-Service (CaaS) offering. The new service provides enterprise AI teams with direct API access to a global network of over 10 million contributors, fundamentally altering the traditional outsourcing model for data collection and annotation. By handing the reins of task design and quality control back to its customers, LXT is making a calculated bid to capture a larger share of the rapidly expanding AI training data market, which is projected to surpass $16 billion by 2032.

This CaaS model allows organizations to integrate a massive human workforce directly into their proprietary systems and MLOps workflows. While LXT manages the sprawling infrastructure of contributor sourcing, payments, and scaling, the customer manages the actual work. It’s a paradigm that directly addresses a persistent frustration among AI developers who need scalable human intelligence without relinquishing control over the intricate details of their data pipelines.

A New Model for AI Data: Control Meets Scale

The core premise of LXT's CaaS is a direct challenge to legacy crowd-sourcing platforms like Amazon Mechanical Turk, where enterprises often face a trade-off between scale and quality control. Unlike passive marketplaces, which can be a black box for data quality and contributor demographics, LXT’s model is built on API integration. This allows AI teams to programmatically deploy and manage data annotation, collection, and evaluation tasks from within their own secure environments.

"AI teams consistently run into the same constraint: they need large volumes of high-quality human input, across many languages, and they need it on their own timeline," said Mark Sewell, VP of Growth at LXT. "With CaaS, they connect our global contributor network directly into their existing workflows via API. They keep full control over how tasks are designed and how quality is measured — we provide the scale. It's a fundamentally different model from traditional outsourcing, and it fits the way modern AI development teams actually work."

This emphasis on control is timely. As AI models, particularly Large Language Models (LLMs) and computer vision systems, become more complex, the demand for highly specific, nuanced, and iterative data labeling has skyrocketed. Industry analysts note that seamless API integration is no longer a luxury but a necessity for efficient AI development, enabling the automation and customization required for sophisticated data pipelines. LXT’s offering, which supports everything from Reinforcement Learning with Human Feedback (RLHF) to search relevance rating, is positioned to capitalize on this demand.

Building a Global Contender Through Strategic Acquisition

LXT’s ability to field a 10-million-strong contributor network did not materialize overnight. The foundation of the CaaS offering is the company's 2024 acquisition and subsequent integration of German-based clickworker, one of the world's largest crowdsourcing providers. The integration, completed in mid-2025, was a decisive move that combined clickworker’s massive global community and self-service platform technology with LXT’s expertise in managed, high-quality data services for enterprise clients.

The acquisition immediately positioned the combined entity as a top-tier competitor to market leaders like Appen. Crucially, it gave LXT a formidable asset: a genuinely global and diverse workforce. This includes a strong contributor base across Europe—a region where many US-centric platforms have historically struggled to gain a foothold. By actively sourcing and managing contributors across more than 150 countries and over 1,000 language locales, the company can offer the demographic and linguistic diversity critical for building less biased, globally relevant AI.

This strategic consolidation underpins the CaaS launch, providing the scale necessary to service the voracious data appetites of modern AI developers. The move also brought clickworker’s former Managing Director, Christian Rozsenich, into the LXT leadership team as Chief Technology Officer, signaling a deep commitment to leveraging the acquired platform's technological strengths for future growth.

Addressing the Enterprise Mandate for Quality and Compliance

Beyond scale and control, LXT is targeting a critical vulnerability in the data-sourcing market: security and compliance. For enterprises operating in regulated industries or handling sensitive information, the freewheeling nature of open crowd marketplaces presents an unacceptable risk. LXT addresses this head-on by certifying its CaaS model under ISO 27001, an international standard for information security management. Furthermore, the platform is fully GDPR compliant, a non-negotiable requirement for any company working with data from European citizens.

This focus on security is complemented by a managed approach to data quality. Instead of leaving quality assurance entirely to the customer, the platform incorporates built-in mechanisms like gold tasks, multi-pass reviews, and real-time analytics. This hybrid approach—customer control backed by platform-enforced quality—aims to provide a more reliable and predictable alternative to the often-inconsistent results from unvetted crowds. By curating a diverse and qualified network, the company also aims to mitigate the algorithmic bias that can arise from training models on homogenous or unrepresentative datasets, a key concern in the push for more responsible AI.

For organizations building the next generation of generative AI, speech recognition, and computer vision applications, this combination of scale, control, and compliance presents a compelling value proposition. It promises to accelerate development cycles by removing the traditional bottlenecks associated with sourcing high-quality, diverse, and secure training data. As AI systems become more integrated into the fabric of business and society, the integrity of the human-generated data that powers them has become a paramount concern for developers and investors alike.