AI at Genio

We believe AI can make learning radically more personalized, effective, and accessible for New Majority Learners.

This guidance is anchored to a learner-first principle, ensuring AI is used to scaffold confidence and skill rather than replacing the necessary cognitive effort.

This is achieved by explicitly removing unproductive friction (like searching or manual busywork) while deliberately preserving productive friction (like refinement and problem-solving) that is essential for genuine learning.

Our Genio AI vision

We believe AI will revolutionize education, making learning radically more:

Personalized: tailored to each learner’s goals, progress, and context.
Efficient: removing wasted time and unproductive friction.
Accessible: supporting the diverse needs of the New Majority Learner.

At the heart of all of this is our ambition to be the learner’s companion and institution’s partner - ethical, pragmatic, and relentlessly focused on outcomes.

To stay aligned, all Genio AI initiatives will follow these anchors:

Learner first: AI should scaffold confidence, agency, and skill, not replace effort or erode independence.
Innovation through application: We will not build foundational large language models. Our strength is applying learning science and proven technologies in unique workflows, interfaces, and experiences.
AI first design: Tools must be reimagined from first principles, remaining beautifully simple. We design for connection, structure, and persistence, not as bolt-ons to legacy workflows.
Ethics by default: No features that bypass learning or increase complexity. AI should always simplify and support, not overwhelm.
Data stewardship: Trust is central. We will balance speed of innovation with responsible handling of learner and institutional data.

AI and the learning process at Genio

Genio’s learning principles

Our approach to AI is directly informed by Genio’s learning principles, which define how we believe learning happens and what learners need to thrive. These principles serve as a constant reference point as we design and implement AI tools, ensuring our innovations support, rather than distort, the learning process:

Heart: Confidence, agency, and enjoyment builds lifelong learners

AI tools should foster learner confidence through clear feedback and encouragement, enhance agency by offering choice and control, and contribute to enjoyment by reducing unproductive friction and supporting meaningful progress.

Mind: Learning is a cognitive process

Our AI-powered tools should improve comprehension and reduce cognitive load. It can help learners develop skills, identify and overcome bottlenecks to their learning, structure their study strategically, and build knowledge as a foundation for deeper thinking.

Action: Learning happens through doing

AI should encourage active output, interaction, and feedback-seeking, not passive consumption. It should support the creation of efficient retrieval activities without replacing the learner’s cognitive effort or participation.

Productive vs. unproductive friction

Not all effort in learning is equal. Some effort is productive, because it strengthens comprehension, retention, and transfer. Other effort is unproductive, because it drains time and attention without advancing learning. Our responsibility is to design AI that removes unproductive friction while preserving and scaffolding productive friction.

Unproductive friction:

Searching: Time wasted finding information or support. Searching is rarely the real skill to be learned.
Copying & context switching: Moving content between tools, juggling multiple systems, or duplicating resources.
Information overload: Being overwhelmed with excessive or poorly-structured content, a common side-effect of AI today.
Unproductive extra processing: Tasks like manually editing transcripts, audio, or formatting that do not deepen understanding.

Productive friction:

Retrieval: The effort of recalling knowledge before being shown the answer.
Application & problem-solving: The struggle to apply knowledge to new contexts, which builds transfer.
Constructive reworking: Purposeful restructuring of notes or concepts that requires the learner to summarise, structure, or reframe in their own words.
Ownership: Learners doing the work of thinking, not outsourcing it entirely to AI.

Genio’s role is to make studying less frustrating and more effective.

This is underpinned through comprehensive understanding of the learning goal and when and where to remove friction, ongoing evaluation to identify where shortcuts emerge to facilitate deeper practice and always asking if there is a net benefit to any new functionality for the learner.

By applying this lens, we can automate boldly while ensuring that learning effort remains where it matters most.

Version history	Publication date
Version 1	July 2026

Read our full AI policy

The table below details how our AI practices align with the SOC 2 Trust Services Criteria, which cover the key areas of security, availability, processing integrity, confidentiality, and privacy.

To learn more about our SOC2 compliance and security practices, head to the Genio Security page.

Criteria	Principle	Objective
Security	Access Control	Protect systems from unauthorized access by implementing role-based access controls, multi-factor authentication (MFA), and regularly reviewing access privileges. This ensures only authorized personnel can access and modify AI systems and data.
	Risk Management	Mitigate vulnerabilities in our AI infrastructure by conducting regular risk assessments, vulnerability scans, and penetration testing. We proactively address identified threats and maintain an incident response plan to handle security breaches.
Availability	System Reliability	Ensure high uptime and continuous service by designing AI systems with redundancy, load balancing, and automated failover mechanisms. We monitor system performance and resource utilization to prevent bottlenecks and ensure scalability.
	Business Continuity	Maintain a tested disaster recovery plan that outlines procedures for data backup and system restoration to minimize service disruption in case of an outage or disaster.
Processing Integrity	Data & Lifecycle Integrity	Ensure data accuracy throughout the AI lifecycle by implementing data validation checks, clear data lineage, and quality assurance at each stage, from data ingestion to model deployment and retirement.
	Performance & Accuracy	Monitor and validate AI system performance against established benchmarks and Key Performance Indicators (KPIs). We document our testing methodologies and regularly audit model outputs to ensure they are complete, accurate, and reliable.
Confidentiality	Data Protection	Safeguard confidential data and intellectual property by encrypting data both in transit and at rest. We enforce strict data handling policies for proprietary algorithms, trade secrets, and internal business information.
	Secure Processing	Process sensitive data securely by using secure computing environments and anonymizing data where possible. All personnel are bound by confidentiality clauses to protect confidential information.
Privacy	Data Minimization	Collect only necessary Personal Data by limiting data collection to what is essential for the function of the AI system. We also implement procedures for data masking and pseudonymization to reduce privacy risks.
	Regulatory Compliance	Adhere to all relevant privacy laws and regulations, including UK GDPR, CCPA, FERPA. We proactively monitor legislative changes to ensure our privacy controls remain compliant and up-to-date.

AI policy template for institutions

We've created a comprehensive AI Policy template, readily available for institutions to download and customize, providing a robust framework for responsible AI implementation.

Download the template