{"id":54220,"date":"2026-08-24T12:20:25","date_gmt":"2026-08-24T11:20:25","guid":{"rendered":"https:\/\/www.intelligentcio.com\/north-america\/?p=54220"},"modified":"2026-08-24T12:20:26","modified_gmt":"2026-08-24T11:20:26","slug":"what-pebls-ai-shift-shows-about-scaling-compliance-heavy-work","status":"publish","type":"post","link":"https:\/\/www.intelligentcio.com\/north-america\/2026\/08\/24\/what-pebls-ai-shift-shows-about-scaling-compliance-heavy-work\/","title":{"rendered":"What Pebl\u2019s AI shift shows about scaling compliance-heavy work"},"content":{"rendered":"\n

Francoise Brougher, CEO, Pebl, on what it takes to turn human expertise into scalable AI.<\/em><\/p>\n\n\n\n

How do you turn a high-touch, expert-led service into a scalable product without breaking the trust customers depend on?<\/p>\n\n\n\n

That question sits behind Pebl\u2019s shift from a managed services provider to an AI-driven company. For more than a decade, Palo Alto based Pebl helped companies hire internationally and manage country-specific requirements across payroll, benefits, onboarding and compliance.<\/p>\n\n\n\n

\u201cCustomers trusted Pebl because specialist teams could navigate complex, context-specific work, from legal questions to custom contracts. But that expertise was difficult to scale. Even routine questions could require multiple emails to clarify context, confirm details and reach the right answer,\u201d said Francoise Brougher, CEO, Pebl<\/p>\n\n\n\n

Pebl\u2019s AI shift focuses on purpose-built AI for global employment. It is designed to help users complete tasks, surface the next relevant step, answer questions when needed and recognise when expert review is required. Routine work can therefore move through the platform faster, while complex or high-risk matters still reach the appropriate specialists.<\/p>\n\n\n\n

Making global employment expertise faster<\/strong><\/p>\n\n\n\n

Pebl started with frequent, context-heavy customer work: compliance guidance, hiring timelines, compensation, onboarding, reporting and support. These requests appeared every day but still demanded accuracy, country context and judgement.<\/p>\n\n\n\n

\u201cAthena was built to help operations teams handle that work faster. Pebl\u2019s internal AI assistant can draw from approved knowledge sources, summarise situations, suggest paths to resolution and help draft customer communications,\u201d said Brougher.<\/p>\n\n\n\n

In a managed services model, the final answer is only part of the work. Teams must clarify context, find appropriate guidance and turn it into a useful response. Athena reduces that friction while keeping specialists accountable for outcomes.<\/p>\n\n\n\n

\u201cThis became Pebl\u2019s first step towards scaling expertise without separating it from human judgement. Teams could accelerate routine or repeatable work, while complex cases remained with the people responsible for reviewing and resolving them,\u201d said Brougher.<\/p>\n\n\n\n

Pebl then applied the same principle to sales through DealPilot, an internal LLM agent providing faster access to approved information on product capabilities, commercial terms, customer-facing answers and deal guidance. Reducing the time representatives spend locating information allows them to concentrate on deal strategy, customer context and conversations that move prospects forward.<\/p>\n\n\n\n

Together, Athena and DealPilot represent the first phase of Pebl\u2019s AI shift: reducing internal friction around expert work. After proving the model internally, Pebl began extending it into the product experience.<\/p>\n\n\n\n

Bringing AI into customer and employee workflows<\/strong><\/p>\n\n\n\n

Built and maintained by Pebl, Alfie is the company\u2019s global employment AI assistant for customer and employee workflows.<\/p>\n\n\n\n

Alfie is purpose-built employment AI rather than a general chat layer. It centres on work customers and employees need to complete, including finding country-specific guidance, generating workforce reports, obtaining support, requesting time off and approving employee actions.<\/p>\n\n\n\n

\u201cThe experience is therefore more action-based than chat-based,\u201d said Brougher.<\/p>\n\n\n\n

\u201cAlfie can answer questions, but its broader purpose is to help users complete tasks inside the platform and surface the next relevant step before they need to search for it.\u201d<\/p>\n\n\n\n

It has also moved from answers into action. Employees supported through Pebl can request time off, while company administrators or PTO managers can approve requests. Pebl is extending the same model into additional workflows, including payroll-related tasks such as bonus creation.<\/p>\n\n\n\n

Building trust into the AI operating model<\/strong><\/p>\n\n\n\n

Pebl\u2019s AI workflows are designed around role-based access control, data protection, human oversight and clear limits on use. Users can access only information they are authorised to see. A manager, for example, may see information relating to a team, while finance users or administrators may access different accounts, payroll or reporting data according to their roles.<\/p>\n\n\n\n

Customer data is protected through strict controls governing AI inputs and outputs. Areas involving legal, tax, compliance or professional judgement remain subject to human review.<\/p>\n\n\n\n

Across Alfie, Athena and DealPilot, AI can retrieve, draft, summarise and structure information. People remain responsible for validation, context, communication and final decisions.<\/p>\n\n\n\n

\u201cThis operating model requires ownership beyond the technology team. Knowledge quality, permissions, escalation, review processes and use-case selection become part of how the business operates. The trust model must function across internal operations, commercial workflows and customer-facing experiences because the same employment expertise now moves through all three,\u201d said Brougher.<\/p>\n\n\n\n

The lesson for scaling AI<\/strong><\/p>\n\n\n\n

Pebl\u2019s transformation offers a broader lesson for compliance-heavy service businesses. The challenge is not simply using AI to move faster. It is determining which elements of an expert-led service can become product workflows, which still demand human judgement and how both can scale without weakening customer trust.<\/p>\n\n\n\n

For CIOs building resilient teams, that balance can support faster hiring in new markets, more consistent support for distributed employees and clearer execution across compliance-heavy workflows. AI provides leverage, but expertise, accountability and trust remain the foundations that make that leverage useful.<\/p>\n\n\n\n

BOX OUT: Three lessons for compliance-heavy AI<\/strong><\/p>\n\n\n\n

First, begin with work that repeatedly pulls experts away from complex issues requiring their attention. Frequent questions around compliance, hiring timelines, onboarding, compensation and support were repeatable enough for Pebl to structure without pretending they were simple.<\/p>\n\n\n\n

Second, turn expertise into workflows rather than merely searchable knowledge. In compliance-heavy work, providing an answer is only part of the value. Systems must surface the next step, capture context, route exceptions and make clear when human review is necessary.<\/p>\n\n\n\n

Third, decide where AI should stop before expanding it. As AI moves into law, payroll, tax, benefits and employee experience, businesses must define what systems can draft or recommend, when people must validate outputs and which decisions require escalation.<\/p>\n","protected":false},"excerpt":{"rendered":"

Francoise Brougher, CEO, Pebl, on what it takes to turn human expertise into scalable AI. How do you turn a high-touch, expert-led service into a scalable product without breaking the trust customers depend on? That question sits behind Pebl\u2019s shift from a managed services provider to an AI-driven company. For more than a decade, Palo […]<\/p>\n","protected":false},"author":58,"featured_media":54221,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[6936,9,43],"tags":[226,12834,153,12835,376,4182,12839,12836,12838,12837,12086,4342,12833,12832,3527],"class_list":["post-54220","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-ai","category-case-studies","category-top-stories","tag-ai","tag-alfie","tag-artificial-intelligence","tag-athena","tag-automation","tag-compliance","tag-compliance-workflows","tag-dealpilot","tag-employment-technology","tag-global-employment","tag-hr-technology","tag-llm","tag-payroll","tag-pebl","tag-workforce-management"],"acf":[],"publishpress_future_workflow_manual_trigger":{"enabledWorkflows":[]},"_links":{"self":[{"href":"https:\/\/www.intelligentcio.com\/north-america\/wp-json\/wp\/v2\/posts\/54220","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.intelligentcio.com\/north-america\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.intelligentcio.com\/north-america\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.intelligentcio.com\/north-america\/wp-json\/wp\/v2\/users\/58"}],"replies":[{"embeddable":true,"href":"https:\/\/www.intelligentcio.com\/north-america\/wp-json\/wp\/v2\/comments?post=54220"}],"version-history":[{"count":1,"href":"https:\/\/www.intelligentcio.com\/north-america\/wp-json\/wp\/v2\/posts\/54220\/revisions"}],"predecessor-version":[{"id":54222,"href":"https:\/\/www.intelligentcio.com\/north-america\/wp-json\/wp\/v2\/posts\/54220\/revisions\/54222"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.intelligentcio.com\/north-america\/wp-json\/wp\/v2\/media\/54221"}],"wp:attachment":[{"href":"https:\/\/www.intelligentcio.com\/north-america\/wp-json\/wp\/v2\/media?parent=54220"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.intelligentcio.com\/north-america\/wp-json\/wp\/v2\/categories?post=54220"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.intelligentcio.com\/north-america\/wp-json\/wp\/v2\/tags?post=54220"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}