Calgary-based LeX Technologies has launched a new AI platform aimed at one of the biggest problems facing businesses experimenting with artificial intelligence: how to use AI on internal knowledge without putting sensitive information at risk.
The company this week announced LeX AI, a patent-pending architecture designed to help enterprises search and interpret internal documents, policies, procedures, contracts, technical files, compliance records, and other business-critical information in a more secure and reliable way.
For many organizations, the promise of AI is obvious. Companies have years of internal knowledge spread across thousands of documents, but finding the right answer can still take hours. At the same time, feeding sensitive information into general-purpose AI tools can create concerns around data exposure, hallucinated answers, inconsistent interpretations, and whether proprietary information could be used to train outside models.
LeX says its platform was built to address those concerns directly.
PRESENTED BY
Rather than relying solely on traditional large language models, LeX AI uses what the company describes as deterministic retrieval and patent-pending reasoning layers. In practice, that means the system is designed to pull answers from trusted source material, cite where those answers came from, and reduce the risk of AI-generated responses that sound confident but are not supported by the underlying documents.
“Businesses should not have to choose between the speed of AI and the security, accuracy, and control required for enterprise use,” said Tristan Vivier, CEO of LeX Technologies (pictured with CTO Imad Victor Lahoud, left, and COO Glenn Sando, right).
“LeX AI was built to make AI-powered research safe for business. Our architecture allows organizations to retrieve precise, cited, and actionable answers from their own knowledge without exposing sensitive data or relying on unreliable AI outputs.”
The company is positioning LeX AI for organizations operating in high-stakes environments where a wrong answer can create real consequences. That includes teams working in compliance, legal, risk, audit, operations, regulatory affairs, and strategic decision-making.
The platform is built to give users source-cited answers from internal knowledge, help teams interpret information more consistently, reduce time spent searching and validating documents, and lower the computing costs often associated with AI-powered research.
“Trust is the missing layer in enterprise AI,” said Prof. Imad Victor Lahoud, Chief Technology Officer at LeX Technologies. “We designed LeX AI around a simple principle: every answer should be accountable to its source. Deterministic retrieval and verification change what businesses can expect from AI research.”
LeX has also developed a secure AI regulatory research tool, designed to provide source-cited answers across complex regulatory environments. The company says that tool can be used on its own or alongside LeX AI, allowing businesses to connect external regulatory intelligence with their own internal policies, procedures, compliance materials, and operating documents.
For companies in regulated or operationally complex industries, that could mean fewer hours spent digging through manuals, standards, regulations, and internal knowledge bases to answer questions that affect day-to-day decisions.
“For operational teams, critical information is often buried across thousands of pages of internal documents, regulations, standards, procedures, and policies,” said Glenn Sando, Chief Product Officer at LeX Technologies. “LeX AI turns that complexity into answers teams can act on immediately, with visibility into the source behind every decision.”
The launch comes as more businesses look for ways to move AI beyond pilots and experiments. While the last few years have been defined by rapid advances in generative AI, adoption inside large organizations has often been slowed by security, accuracy, governance, and trust concerns.
LeX is betting that the next phase of enterprise AI will be less about bigger models and more about safer architecture.
“AI adoption will not be defined only by who has the most powerful model,” Vivier added. “It will be defined by who can make AI safe, secure, accurate, and useful inside real businesses. That is what LeX is building.”
LeX AI is currently available for enterprise organizations, with the Calgary company working with select partners across regulated and operationally complex industries on initial deployments.
Leave a Reply