top of page
Insights

Market Insight & Strategic Intelligence

Seer by Workvivo and the EU AI Act Compliance Tightrope

  • Writer: Tim Banting
    Tim Banting
  • May 26
  • 3 min read

Workvivo by Zoom has officially launched Seer, a standalone ‘People Intelligence’ platform designed to transform employee feedback into predictive insights. The rollout, led by industry veteran Justin Black, comes at a critical juncture as the European Union’s landmark AI regulations reach significant milestones for high-risk workplace systems.


Seer aims to bridge the ‘execution gap’ by combining traditional employee surveys with real-time ‘signals’ from collaboration tools. While Workvivo’s launch focuses on improving engagement and manager effectiveness, the platform’s core functionality (monitoring and evaluating employee behaviour), places it within the High-Risk category under Annex III (Point 4b) of the EU AI Act. This classification triggers stringent compliance mandates regarding transparency, data governance, and human oversight.

What: Seer’s Strategic Position and EU AI Act Compliance Requirements


The employee engagement sector is undergoing a fundamental pivot from annual, static surveys to continuous, automated ‘listening.’ Seer enters this space as a direct challenger to established players like Qualtrics and Glint. Its competitive edge lies in deep integration with the Zoom/Workvivo ecosystem, allowing it to surface organisational risks and themes in real-time across global workforces.


As of May 2026, EU AI Act compliance has become a primary hurdle for such platforms. Because Seer is used to monitor and evaluate worker performance, it must adhere to the EU’s latest high-risk guidelines. While certain prohibited practices (such as biometric emotion recognition), were banned in 2025, the timeline for standalone high-risk systems has shifted. Following recent Omnibus legislative updates, the deadline for full compliance with Annex III requirements has been extended to 2 December 2027.


Workvivo currently balances these capabilities with a focus on privacy. The platform enforces a five-response threshold to maintain anonymity in reporting. To comply with the Article 5 prohibition on emotion recognition in the workplace, Seer is designed to limit its analysis to aggregate thematic sentiment from written text. Crucially, the platform must ensure it does not attempt to infer the internal emotional states of specific individuals, a practice banned under the AI Act regardless of whether the input is biometric or textual.


As Seer is classified as a high-risk AI system under Annex III of the EU AI Act, Workvivo is required to formalise comprehensive Annex IV technical documentation. This documentation acts as the mandatory ‘paper trail’, detailing risk management protocols and algorithmic logic, essential for securing its CE marking and ensuring long-term legal standing within the European market.

Capabilities & Limitations


Capabilities

  • Signal Integration: Merges active survey feedback with passive ‘signals’ like engagement patterns and adoption metrics from the Workvivo/Zoom ecosystem.

  • Manager-First Guidance: Provides line managers with personalised dashboards and AI-guided ‘next steps’ to drive morale and retention at the team level.

  • Built-in Action Tools: Directly integrates with Workvivo communication features (Updates, Spaces, Livestreams) to ‘close the loop’ on feedback without leaving the platform.


Limitations

  • Compliance Transparency Gap: Limited public documentation currently exists detailing how the platform satisfies the comprehensive nine-section Annex IV technical package required for high-risk systems.

  • The Privacy Tension: Real-time signal analysis remains a high-scrutiny area for European Works Councils, who often view ‘continuous listening’ as a form of workplace surveillance.

Signals to Watch


  • Conformity Disclosures: Watch for Workvivo’s release of formal ‘Instructions for Use’ (Article 13). These are legally required to help managers avoid ‘automation bias’—blindly trusting AI recommendations over human judgment.

  • The Right to Explanation: Under Article 86, EU employees can now demand an explanation for AI-assisted decisions. Seer’s success depends on its ‘interpretability’: whether it can show the logic behind a ‘risk’ flag or a ‘retention’ score.

  • Enforcement Thresholds: While the full Annex III deadline is now December 2027, the Article 50 transparency requirements (disclosing when content is AI-generated) are still expected to become applicable by 2 August 2026.


Sources

bottom of page