What are you using to filter/monitor Chatgpt for example?

[sigma]

I asked ChatGPT and this is what it told me:

 

You said:

how do I filter and monitor chatgpt in a school?

ChatGPT said:

 

Filtering and monitoring ChatGPT (or any AI tool) in a school setting is important to ensure that its usage aligns with educational goals, maintains safety, and adheres to appropriate content standards. Here are several strategies to filter and monitor ChatGPT effectively:

1. Set Clear Usage Policies

 

Establish Guidelines: Create a clear set of guidelines for students and teachers on how ChatGPT should be used, specifying what constitutes appropriate and inappropriate uses. For example, students should use it as a tool for learning and creativity, not for cheating or bypassing educational tasks.

Terms of Use: Include terms of use that align with the school’s code of conduct, ensuring that students understand the ethical and responsible use of AI tools.

 

2. Restrict Access to ChatGPT

 

Network Restrictions: You can use network-level filtering tools to block access to ChatGPT on school networks. This can prevent students from accessing the service outside of monitored settings.

Whitelist/Blacklist Software: Some software allows you to specify which websites or services are accessible (whitelisting) or blocked (blacklisting). Schools can apply these tools to restrict ChatGPT access based on educational needs.

 

3. Use Monitoring Software

 

Activity Tracking: Use activity monitoring tools to track what students are doing on their devices. You can monitor both browser activity and direct interactions with AI tools like ChatGPT.

Screen Monitoring: Software solutions, such as GoGuardian, LanSchool, or others, can enable teachers and administrators to monitor the screen activity of students in real-time. These tools help detect any unauthorized use of ChatGPT.

Logging Conversations: Consider using a system that logs conversations with ChatGPT to review interactions later. This ensures that any inappropriate behavior or misuse is caught promptly.

 

4. Incorporate ChatGPT in Structured Activities

 

Supervised Usage: Instead of unrestricted access, incorporate ChatGPT into structured, teacher-guided lessons. This can involve using ChatGPT to help with research, creative writing, problem-solving, or brainstorming sessions.

Review Outputs: Teachers can review students' interactions with ChatGPT to ensure the AI’s outputs align with academic goals. Encouraging students to share their ChatGPT responses in class or in assignments ensures oversight.

 

5. Content Filtering

 

AI Filters: Some AI platforms (including ChatGPT) offer built-in content filtering tools that can prevent the AI from generating harmful or inappropriate content. Make sure these settings are activated and adjusted for age-appropriate content.

Third-Party Tools: For even greater control, third-party content filters, like those available in some firewall or internet security systems, can help filter out inappropriate or off-topic content generated by the AI.

 

6. Educate Students About Ethical Use

 

Critical Thinking: Teach students to critically assess the information generated by ChatGPT. Remind them that while AI can be a helpful tool, it may not always provide accurate or reliable information.

Avoiding Plagiarism: Educate students about the potential for unintentional plagiarism when using AI-generated text. Discuss how to properly cite sources and use AI-generated content as a starting point for their work, not the final product.

 

7. Access Controls for Teachers

 

Teacher-Only Access: Allow only educators or administrators to have unrestricted access to ChatGPT. Students can use it under supervision, but teachers may need full access for educational purposes.

Role-Based Permissions: Set up systems where only teachers have the permissions to unlock certain features of ChatGPT, such as accessing advanced capabilities or reviewing detailed interactions.

 

8. Evaluate Use Regularly

 

Audit and Evaluate: Periodically assess how ChatGPT is being used in the classroom to ensure it is supporting learning effectively. Gather feedback from students, teachers, and parents to identify any potential issues with misuse or inappropriate content.

Adjust Policies: Based on evaluations, make necessary adjustme

 

Whilst accurate, its a bit vague.

 

So how do you do this, and how are you planning to handle images, malicious code creation etc in future?

Edited February 14, 2025 by sigma

[Cazab]

Blocked.

[TronXP]

Blocked for students, but they are allowed to use Securly Chat which is monitored and filtered. Staff have access to AI but we have a AI Policy that they must adhere to.

[Disease]

Blocked for students, but they are allowed to use Securly Chat which is monitored and filtered. Staff have access to AI but we have a AI Policy that they must adhere to.

 

Are you willing to share you AI policy via PM, it's something I have to look into and would good to have something to start with?

[sigma]

We are blocked for pupils now, but envisage that it won't be long before we are asked to allow it.

[sigma]

Are you willing to share you AI policy via PM, it's something I have to look into and would good to have something to start with?

 

Same here.....?

[Ditto]

This is a genuine response, but have you posted that question to ChatGPT?

 

Interestingly ChatGPT does have level of self-awareness. I once asked it to analyse an answer to assess it's view on whether it had been AI generated. Fundamentally, it said yes, it effectively if admitted it's own tendency for being verbose! It might also admit to providing some answer with vagueness. For policy advise, I think that's understandable, but when pushed, it can come up with clear structured steps to achieve solutions. Over time, like traditional internet searching, you learn skills to get nearer to what you are looking for.

[sigma]

I asked, it said:

 

what should be included in a acceptable use policy for ChatGPT in a school?

ChatGPT said:

 

An Acceptable Use Policy (AUP) for ChatGPT in a school setting should establish clear guidelines to ensure responsible and ethical use of AI tools. Key components to include are:

 

Purpose and Scope: Define the objectives of using ChatGPT, such as enhancing learning experiences, providing personalized support, and assisting with research. Specify that the policy applies to all students, staff, and faculty members.

 

Permissible Uses: Outline acceptable applications of ChatGPT, including:

Assisting with homework and assignments.

Facilitating language learning and practice.

Providing explanations of complex topics.

Supporting creative writing and brainstorming sessions.

 

Prohibited Uses: Clearly state activities that are not permitted, such as:

Submitting AI-generated content as one's own work without proper attribution.

Using ChatGPT to generate or disseminate false, misleading, or harmful information.

Engaging in plagiarism or academic dishonesty.

Sharing personal, confidential, or sensitive information with ChatGPT.

 

Academic Integrity: Emphasize the importance of honesty in academic work. Specify that any use of AI tools must be properly documented and credited. For instance, if ChatGPT is used to generate text, it should be cited appropriately, similar to other reference materials.

niu.edu

 

Privacy and Data Security: Instruct users not to input personal or confidential information into ChatGPT to protect privacy and comply with data protection regulations.

ekaru.com

 

Monitoring and Compliance: Inform users that their interactions with ChatGPT may be monitored to ensure adherence to the policy. Outline the consequences of policy violations, which may include disciplinary actions.

 

Training and Support: Offer training sessions to educate users on the ethical use of AI tools and provide resources for responsible engagement with ChatGPT.

 

Review and Updates: State that the policy will be regularly reviewed and updated to keep pace with technological advancements and evolving educational needs.

 

By incorporating these elements, the AUP will promote a responsible and effective integration of ChatGPT into the educational environment.

[psydii]

Microsoft Purview communication compliance keeps an eye on Copilot for us.

[Ronnie_Securly]

Securly's AI Chat for Filter can help you here! Give me a shout if you want to see what we can do.