What is Gemini’s ‘Guided Learning’ and ChatGPT’s ‘Study mode’?

For many teachers, AI tools risk undermining independent learning rather than strengthening it.

That’s where features such as ChatGPT's ‘Study mode’ and Gemini's ‘Guided Learning’ come in. Instead of simply delivering responses, these tools are designed to slow students down and guide them through learning processes step by step.

Used thoughtfully, the tools could mark a shift away from ‘AI as an answer machine’ and towards ‘AI as a study partner’. In the sections below, we’ll look at what the tools actually do, how they work and how you might use them in ways that keep students actively learning.

How do the tools work?

Gemini's Guided Learning

Gemini’s Guided Learning is designed to act like a tutor instead of a search engine. When students enter a question or query into the interface, it doesn’t jump right into the answer. Instead, the tool breaks the topic into manageable steps, asks guiding questions and encourages students to explain their thinking before moving on.

It might prompt students with questions such as ‘What do you already know about this?’ or ‘Let’s tackle this one step at a time’. If students get stuck, it offers hints to mirror the scaffolding and guiding questions that many teachers use.

ChatGPT's Study mode

ChatGPT’s Study mode works in a similar way – focusing on learning rather than output. When students enter a prompt, the tool provides some information and study tips to aid learning, and then checks that students really understand before revealing more details.

From a teaching perspective, the ChatGPT study mode feature helps frame AI as part of revision or homework. You could ask students to screenshot their interactions and reflect on what they learned and not just what they produced.

Key similarities

Both tools put the learning process ahead of the final answer. They encourage independent thinking, reasoning and problem-solving rather than bypassing them. Crucially, both features can be toggled on or off, giving students and teachers control over how and when they’re used.

How do these features support educators?

1. Allow differentiation

In many classrooms, there are different levels of learning confidence, prior knowledge and pace. Guided AI features can help smooth that out, so students can move through content without feeling held back or rushed.

Have students use these tools during independent work to revisit a concept they didn’t quite grasp, while others push further with extension questions. In both tools, the responses adapt to students’ prompts, making them suitable for mixed-ability groups.

2. Extend support beyond the classroom

One-to-one support is always limited by time in most classrooms. When students are revising or stuck on homework, guided modes can step in to support them with questions, hints and structures.

Suggest students use Study mode to talk through a problem before class, then come with specific questions. This frees up class time for addressing gaps rather than reteaching from scratch.

3. Build self-efficacy

Because these tools don’t automatically hand over the answers, students are nudged to try, fail safely and try again. Over time, this helps students build confidence in their own thinking.

Ask students what strategies helped, where they got stuck and how they overcame frustrations. This shifts AI use from ‘Did you get the right answer?’ to ‘How did you learn?’ and prioritises effort and reflection.

4. Other modes of support

Beyond core subject content, Guided Learning and Study mode can also support revision planning and exam preparation. Students can practise explaining ideas in their own words or break study tasks into manageable chunks. This can turn into extra academic support to reinforce the learning that already takes place.

How can we integrate these features into study?

Show the difference

Do a live comparison with a question you’d normally set for homework or revision. Run it through Gemini without Guided Learning and then with Guided Learning (and the same for Chat GPT with and without Study mode). You could do this with a tricky maths problem or a short essay prompt and ask students to spot the differences. Which version pushes them to think through the process? Which one feels quicker but less useful for future understanding?

Be clear about when each mode makes sense

Not all tasks require Guided Learning. Sometimes students just need to check a definition or confirm a date. At other times, when planning writing or revising for exams, slowing down is exactly what’s needed.

Have a discussion about when Study mode is the better choice, and when a direct answer from a search is acceptable. This will help students make intentional decisions instead of just going for convenience in their learning.

Co-create class agreements around AI use

Instead of imposing rules, involve students in setting a simple class agreement covering expectations such as:

• Using guided modes for breaking down complex tasks or checking understanding
• Not using AI for assessed work
• Being able to explain what they asked, learned and how it helped them, even if AI was involved.

When students take part in setting these boundaries, they’re more likely to respect them.

Limits to be aware of

As promising as guided AI tools are, they’re not a magic fix. As with any classroom resource, they work best when used with awareness and restraint. Consider the following:

Start with where students are

Before introducing these tools, find out how students already use AI and how they feel about it. Some will be heavily reliant on it, while others may be anxious or unsure of whether they’re allowed to use it at all.

Watch for over-reliance

Although these guided modes slow things down, they can still become a crutch if students lean on them too much. If every problem is mediated through AI, students may stop trusting their own judgement. Counter this by building in study moments where students must first attempt a task without AI.

Human interaction still matters

No AI can replace discussion and feedback from teachers or peers. Learning is social and contextual, and these human elements allow for richer conversations that shouldn’t be replaced. Have students explore a concept with guided prompts at home, then debate or further extend the topic together in class.

Equity and accessibility

Not all students have equal access to devices, reliable internet or premium AI features. Frame these tools as supplementary where possible, to ensure core learning can still take place without them. Transparency about optional use keeps learning fair and inclusive.

Answers don’t mean learning

Perhaps the most important distinction to reinforce is that getting an answer is not the same as learning. Learning involves struggle, reflection and revision, which guided AI features can support only if students engage with them actively.

A smarter step forward for AI in the classroom

Features like Gemini’s Guided Learning and ChatGPT’s Study mode represent a positive shift in how AI can support learning. Rather than race students to the learning finish line, they refocus attention on thinking, reasoning and understanding.

That said, no tool replaces quality teaching. AI works best when it complements strong instruction, clear expectations and meaningful classroom relationships. Guided modes can reinforce the habits and modern skills that educators are building in students.

Further reading

Learn more about safe and effective AI in education. Read Extending school chatbot policies beyond the classroom: Safe-use tips for students, How to use AI to fuel students’ imagination and 5 ways to discourage your students from cheating with AI.

Big Think webinar

Want to learn more? Register to attend or watch on demand our Big Think webinar: Assessment evolved: Redefining formative assessment in a Generative AI era. Meredith Reeve and Muireann Hendriksen, Pearson lead specialists in strategy and research, explore how to evolve formative assessment for an AI-driven future.