Inside classroom.cloud’s safeguarding technology
What it does
Our safeguarding technology keeps your school’s online environment safe by monitoring concerning activity, identifying students at risk and spotting online safety trends.
How we do it
The system looks out for words that students type, copy or paste. When those words or phrases match with our database of safeguarding terms, it triggers an alert.
(Our database has been built over several years with partners such as the Internet Watch Foundation and input from schools to ensure we can carry out a comprehensive level of online safeguarding.)
What we capture
The alert is recorded in our console and that makes it easy for safeguarding staff to assess.
The system can also capture a screenshot or a video to show context in the most serious alerts. Your school can choose the terms this applies to. Alternatively, if desired, you can move keywords between categories to ensure no screen capture is triggered.
How we assign possible risk
For example, in classroom.cloud, the system monitors for safeguarding terms in these categories:
| Adult | Gambling | Radicalisation |
| Bullying | Grooming | Self-harm |
| Cybersecurity | Homophobic | Suicide |
| Drugs | Other | Wellbeing |
| Eating disorders | Racism |
We prioritise what we believe is a potential risk using our contextual intelligence risk index formula. So, when a term from one of those categories is triggered, additional variables are considered:
- whether the student is a member of a particular safeguarding group on the system, which may highlight additional consideration is needed (scores are assigned for: yes or no)
- when the trigger happens (scores are assigned for: inside or out of school hours)
- whether the alert was triggered on a supervised or unsupervised device, e.g., in the classroom with the teacher or personal study time in the library (scores are assigned for: yes or no)
- whether the student has triggered similar word or phrase alerts previously (scores are assigned for: no; more than 5 times in the last month; more than 5 times in the last week; more than 3 times in the last day)
- whether the student has triggered similar category alerts previously (scores are assigned for: no; more than 5 times in the last month; more than 5 times in the last week; more than 3 times in the last day)
- the priority level of the word/phrase matched (scores are assigned for: low; medium; high; urgent)
- whether it was triggered on a high-risk app such as a browser or messenger (scores are assigned for: yes or no)
- whether it was triggered on a website (scores are assigned for: no; yes, any website; medium-risk website; high-risk website)
Once all these variables are considered and scored individually, the system puts them all together to generate the final risk score total, so that safeguarding staff can immediately see the alerts that require urgent attention.
Now, we can see that if a student types a high-risk term such as ‘bomb’ in a spreadsheet in class, it will generate a different score than typing it in a chat room, unsupervised. The scoring, and any changes the school makes to how it is calculated, help to provide the important context, allowing professionals such as Designated Safeguarding Leads to have a fuller picture of what activities take place on the varied devices at the school.
Although our expertise in safeguarding allows us to build this algorithm and set out the suggested scoring, ultimately, your school can alter and tweak this so it fits with your learners’ needs and your approaches to risk. You can also set some websites to never be monitored – for example, HR systems – so words are not triggered out of the student context.
As with all our systems, your school will always be the Data Controller and make the decisions.
Where data is located
All of classroom.cloud’s safeguarding data is stored in NetSupport’s Azure cloud. Your school has control over the data storage settings.
Learn more about managing privacy and personal data in classroom.cloud.
How long we keep data for
We set a default retention period of 13 months to allow for a rolling backup of your data and for your school to ensure you have accurate records across a school year.
Some schools may opt for shorter periods, with designated members of staff (e.g., the DSL) making verified support requests to delete data which has already been dealt with or is no longer needed. Again, this is an important decision that is made solely by the school.
Data processing summary
Your school can choose the thresholds of the data that is captured, for example, the topics you want to include or exclude from monitoring, the sensitivity level applied, the times of day that monitoring is active, and even limiting the location where safeguarding monitoring can take place.
Your school is both the Data Controller and the Data Processor for all of these options.
We aim to ensure that transparency is available through our Data Processing Agreement. By providing guidance on expected and designed usage, we give your school the scaffolds to help keep your staff, students and wider community abreast of how data is processed by NetSupport under your instructions. Simple graphics and readable language are at the heart of this.
Data you provide
Data users provide
Data we collect
Who has access
Your school also controls who has access to your safeguarding data.
For extra data security, our safeguarding module can be used exclusively by safeguarding staff, without the involvement of the IT team. It is designed this way for maximum data privacy and in line with KCSIE guidance.
We openly discuss the impact that this can have on how a school operates in our Privacy by Design and by Default guidance. This is vital for schools to understand how they can control access to personal data – and even sensitive information including Special Category Personal Data.
Who controls all of the above?
Your school controls all of the safeguarding module options – even down to the safeguarding categories you wish to monitor.
Data sharing
We categorically do not share any of your school data.
There are no hidden ‘opt out’ clauses for schools to be concerned about. Your data is in your control, always.
For classroom.cloud, we do have a range of integrations. These are where you, as Data Controller, instruct us to work and transfer data to other Data Processors who you have similar agreements with. These integrations allow schools to securely transfer details such as behaviour management records, allow schools to automate the users and groups based on tools from Microsoft, Google, Clever, Classlink, CPOMS and Tes MyConcern. All integrations are those set out by the school and under your instructions.
What sets classroom.cloud apart?
Transparency
NetSupport makes its list of keyword phrases visible to schools. Most solutions either do not or may lack the transparency we strive to provide.
Keeping humans in the loop
When dealing with children’s safeguarding, we believe in the value of human oversight.
NetSupport does not – and has never – used AI in its safeguarding tools. If we left such sensitive decision-making to technology, it could mean that a student may wrongly have a concern raised against them, when human oversight could easily have concluded otherwise.
External validation
classroom.cloud has been externally reviewed and validated by Education Alliance Finland. It has also gained the ISTE Seal of quality for educational tools.
NetSupport is also ISO27001 certified for its commitment to data security and routinely matches this against standards from NIST and other framework providers.
We make use of open source templates for sharing technical and security information and continue to look for other ways we can demonstrate our accountability.