Online English language testing for employment: Is it secure?

Jennifer Manning
Woman with a headset at a computer

Managers and HR professionals have a global workforce at their fingertips – and now, . This makes adopting a secure English language test for employment more important than ever.

An online English test enables organizations to assess candidates’ language proficiency from anywhere in the world, screen more applicants, and standardize the hiring process. They also help HR professionals and managers to save time – ensuring only people with the right language skills advance to the interview stage.

But how can employers be certain these tests are safe? And how easy is it for people to cheat? In this article, we’ll explore a few of the top security concerns we hear, and share what features make online language tests secure.

What is an online English test?

An online English test measures how well a job applicant can communicate in English, focusing on speaking, listening, reading and writing skills. They also assess a candidate’s specific English for business skills – for example, how clearly someone can communicate on the phone with clients, or understand what is being said during a conference call.

Online tests can be taken in a controlled environment – in a testing center with in-person proctors – but also from a job applicant’s personal computer or mobile phone at home. When tests are taken at home, they can be made more secure using virtual proctors or powerful AI monitoring technology.

Cheating, grading and data security

When many people think of taking a language test, they imagine the traditional way: students in a large testing center scribbling away with pen and paper. No mobile phones are allowed, and if test-takers are caught cheating, they’ll be flagged by a proctor walking around the room.

So when managers or HR professionals consider the option of an online English test – taken digitally and often without human supervision – it’s no surprise that many have questions about security. Let’s take a look at some common concerns:

Is cheating a problem?

A large number of test takers admit to cheating on their tests. According to research by the International Center for Academic Integrity, 68% of undergraduate students say they’ve cheated on a writing assignment or test, while 43% of graduate students say they have.

But how easy is it to cheat during a Versant test?

The truth is, not very. With Versant, exam cheating is actually quite difficult, and test takers would have to outsmart a range of AI monitoring technologies.

If a verified photo is uploaded to the platform, HirePro’s face recognition technology can compare the live test taker with it. This ensures test takers are who they say they are, and haven’t asked someone else to sit the exam for them. It is the institution’s responsibility to verify the original photo.

And since Versant tests are monitored using specialized AI algorithms – without a human present – even the slightest suspicious behaviors are flagged for review. For example, Versant notices if a different face appears in the video, or if the camera goes dark. With video monitoring, our platform also flags if the test taker moves from the camera, or looks away multiple times. And we’ll see if someone changes tabs on their computer.

Finally, the entire test is recorded. When suspicious behavior arises, HR professionals will decide whether to accept or reject the results – or have the candidate retake the test.

Are scores accurate?

We’ve all had frustrating experiences with AI. Chatbots don’t always understand what we’re trying to say, and speech recognition technology sometimes isn’t up to par. This leaves many wondering if they should trust AI to grade high-stakes tests – especially when the results could be the difference between someone getting the job, or not.

Versant uses patented AI technology to grade tests that are trained and optimized for evaluating English language proficiency. It evaluates speaking, listening, reading, writing, and even intelligibility.

Our AI is trained using thousands of fluent and second-language English speakers. With these models, we’re able to not only evaluate how someone should be assessed but also understand when they’ve mispronounced words or have made another mistake. Using all this information, a candidate’s final score is evaluated based on more than 2000 data points.

Do online tests follow GDPR standards?

HR professionals and managers deal with sensitive personal information every day. This includes each job applicant’s name, full address, date of birth, and sometimes even their social security number. The HR tools they implement therefore must also keep this data secure.

Most importantly, it must follow GDPR standards. The data must be gathered with consent and protected from exploitation. With Versant, test-taker data is securely stored and follows all .

All our data is encrypted at rest and in transmission. Versant assessment data is stored in the US and HirePro, our remote monitoring partner, stores the proctoring data in either Singapore or Europe, depending on customer needs. Both systems are GDPR compliant.

Versant: a secure English language test

ճ Versant automated language test is powered by patented AI technology to ensure the most accurate results for test takers and employers alike. Even better, our remote testing lets HR professionals securely and efficiently assess candidates worldwide, 24/7 – and recruit top global talent to help more companies scale.

More blogs from app

  • Children engaged in a classroom activity, with colorful educational posters and a banner in the background.

    How to use flipped learning to support your learners

    By
    Reading time: 6 minutes

    What is flipped learning?

    To understand better what flipped learning is, first let’s see how it differs from blended learning, a term with which it is often confused.

    Blended learning is a way of teaching that combines face-to-face classroom teaching with online resources. We freely use online resources to create a more personalized learning experience.

    Flipped learning is a little different. As the name suggests, it "flips" a traditional lesson. It tells us exactly which stages of the lesson should go online. In a flipped learning class, all of the more traditional aspects (also called "study stages"), are completed online and the homework (also called "application stages") comes into the classroom.

    Let’s look at an example.

    A typical receptive skills lesson normally has six stages:

    1. Lead in
    2. Set context
    3. Pre-teach vocabulary
    4. Gist task
    5. Detailed task
    6. Follow up

    With a traditional teaching model, we do the first five stages in class and set the last one for homework. With flipped learning there are a few ways to tackle these stages, but a basic model would look like this:

  • Precision teaching with AI: Aligning GSE objectives with generative AI for targeted materials

    By
    Reading time: 4 minutes

    English teachers today face increasing demands: create engaging content, differentiate instruction and address diverse learner needs – all within a limited time. The rise of Generative AI, like ChatGPT, offers a promising solution. But without proper guidance, AI-generated content can lack educational value. This blog post introduces a practical, research-informed approach to using AI tools aligned with the Global Scale of English (GSE). You will learn how this framework helps educators design accurate, personalized and level-appropriate English teaching materials quickly and confidently.

    Why GSE and AI are a game-changing combination for ELT

    The Global Scale of English (GSE) is a CEFR-aligned framework developed by app, offering detailed "can-do" learning objectives. It includes nearly 4,000 descriptors across speaking, listening, reading and writing skills, offering more precision than traditional level labels like A2 or B1. At the same time, Generative AI tools such as ChatGPT can generate entire lessons, tasks and assessments in seconds. The challenge lies in ensuring this content is aligned with clear pedagogical outcomes.

    Pairing AI’s creative speed with the GSE’s structured outcomes offers a scalable way to meet learner needs without compromising instructional quality.

    Unlocking measurable, differentiated and efficient teaching with GSE and AI

    The GSE makes objectives measurable

    Unlike generic teaching goals, GSE objectives are specific and measurable. For example, a B1-level learner objective might state:

    “Can identify a simple chronological sequence in a recorded narrative or dialogue.” (GSE 43)
    This clarity helps teachers define outcomes and ensure each AI-generated task targets an actual language skill, not just generic content.

    Generative AI enhances productivity

    Teachers using Generative AI can create draft lesson materials in minutes. By inputing a structured prompt such as:

    “Create a B1 reading activity that helps learners summarize the main points of a short article.”
    ChatGPT can instantly generate content that meets the learning goal. When guided by the GSE, AI becomes a collaborative assistant as well as a time-saver.

    The GSE + AI combination supports differentiation

    Because the GSE includes descriptors across a wide proficiency range (from pre-A1 to C2), teachers can tailor AI-generated content to meet the exact needs of their students. Mixed-level classrooms or tutoring contexts benefit especially from this, as teachers can create multiple versions of a task with consistent scaffolding.

    Practical tips

    • Use the GSE Teacher Toolkit to select objectives based on skill, level or function.
    • When prompting ChatGPT, include the GSE descriptor in your input for more precise results.
    • Always review and adapt the AI output to match your learners’ context, culture and curriculum.
    • Create a prompt library mapped to GSE codes to save time in future planning.

    A step-by-step example of the GSE and AI in action

    Here is a typical application of the workflow:

    1. A teacher selects a GSE objective, such as:
      “Can write a basic formal email/letter requesting information.” (GSE 46).
    2. Within seconds, a sample formal email, accompanied by a short reading comprehension task and a vocabulary activity, is generated.
    3. The reading task serves as a model to help learners analyze the structure, tone, and key language features of a well-written email before attempting their own.
    4. The teacher then reviews and refines the output for clarity, appropriateness, and context relevance.

    This process supports targeted teaching while significantly reducing preparation time.

    Overcoming challenges: Ensuring quality and relevance

    Challenge: AI outputs may lack cultural context, level appropriateness or instructional clarity.
    Solution: Always pair AI with professional judgment. Use the GSE to check that skills match the intended outcome, and adjust the complexity of the language as needed.

    Challenge: Teachers may be unfamiliar with how to write effective AI prompts.
    Solution: Start simple with templates like:

    “Create a [skill] activity at [level] that supports this GSE objective: [insert objective].”

    Challenge: Risk of over-relying on AI for instruction.
    Solution: Use AI as a starting point, not the final product. Combine AI-generated content with classroom interaction, feedback and your own creativity.

    Teaching tools that make this easier

    • : for exploring and selecting level-appropriate learning objectives
    • : for generating customizable teaching content
    • GSE Smart Lesson Generator: an AI-powered lesson creation tool developed by app that uses the GSE framework to automatically generate high-quality activities and lesson plans
    • Google Docs or Word: for editing and organizing your materials before class

    Confidently transforming English teaching

    Combining Generative AI with the Global Scale of English allows teachers to design materials that are both fast and focused. The GSE provides the structure; AI provides the speed. Together, they offer a sustainable solution for personalized English instruction that respects both learner needs and instructional quality.

  • Teacher and young students engaging in a classroom activity, surrounded by colorful art supplies and educational materials.

    Back-to-school challenges that teachers face – and how to solve them

    By
    Reading time: 4 minutes

    A new school year brings excitement and opportunity. It also presents challenges for teachers, regardless of their level of experience. Here are common problems that teachers face and simple tips to help you start the year with confidence.

    1. How do I establish effective classroom management and routines?

    The challenge:
    A new year means a new group of students, each with unique personalities and expectations. Setting clear routines and managing classroom behavior can be daunting, especially when students test boundaries or struggle to adjust.

    The solution:
    Set the tone early by communicating clear expectations for behavior, participation and respect. Use visual aids and consistent language to reinforce routines. Involve students in rule-setting to foster ownership and accountability. Consistency and patience are key; routines may take time to solidify, but your commitment will pay off.

    2. How can I build strong relationships with my students?

    The challenge:
    Connecting with a diverse group of learners – some of whom might be eager, while others might be more reserved – can be challenging. Building trust and rapport is essential for engagement and academic success, but it doesn’t happen overnight.

    The solution:
    Dedicate time to getting to know your students as individuals. Use icebreakers, interest surveys and daily check-ins to show you care about their lives beyond academics. Incorporate culturally-responsive teaching practices to respect students’ backgrounds and experiences. Even small gestures, such as greeting students by name and celebrating their achievements, can make a big difference.