A plagiarism check in Brightspace is an automated comparison that runs when a student turns in written work. The moment a file lands in an assignment folder, the platform can hand that text to an originality service, which scans it against a large body of sources and returns a similarity report. The instructor sees a percentage and a color-coded breakdown of matched passages. That is the whole idea in one sentence. Everything else in this guide is just the detail of how that sentence plays out for the three people who touch it: the student who submits, the instructor who reviews, and the administrator who wires it all together.
Brightspace, made by D2L, is one of the more widely used learning management systems in schools and universities. If you have used Canvas, Moodle, or Blackboard, the shape will feel familiar. The vocabulary is slightly different, though, and small differences in vocabulary are exactly where people get stuck. So let us walk through the actual workflow rather than the marketing version of it.
What a "plagiarism check" actually is
It helps to be precise, because the phrase hides two very different things.
The first is a similarity check. This compares a submitted document against web pages, published works, and often a database of previously submitted student papers. It does not have an opinion about whether something is cheating. It simply reports overlap: this sentence appears here, that paragraph matches a source there. A similarity score is a measurement, not a verdict.
The second is AI writing detection, which is newer and works on a completely different principle. Instead of matching text against existing sources, it estimates the statistical likelihood that a passage was generated by a language model. A sentence can be entirely original, match nothing on the web, and still get flagged as likely AI-written. These are separate questions, and treating a single percentage as the answer to both is the most common mistake we see.
Brightspace itself does not invent these checks. It provides the plumbing. The actual analysis comes from an integrated tool that plugs into the assignment workflow. Understanding that division of labor explains almost everything else.
How it works: the student's path
For a student, the originality check is mostly invisible, which is how it should be. Here is the sequence.
A student opens an assignment, which in Brightspace lives in a feature historically called the Assignments tool (older versions and longtime instructors still call individual assignments "dropboxes," and you will hear both words). They attach a file, type a comment if the instructor asked for one, and click submit.
At submission, a few things happen at once. Brightspace records a timestamp, stores the file, and, if originality checking is enabled for that assignment, passes the document to the connected service. A confirmation appears, and many institutions send a confirmation email as well. That confirmation is worth keeping. If there is ever a dispute about whether something was turned in on time, the submission receipt is the cleanest evidence available.
Depending on how the instructor configured things, the student may or may not see the similarity report themselves. Some courses let students view their own originality score before the deadline so they can revise and resubmit. This is one of the most useful settings in the entire system, because it turns the check from a trap into a teaching tool. A student who sees that a paragraph is flagged can go back, rethink the phrasing, and learn what proper paraphrasing and citation actually look like.
A few practical notes that save students real grief. File format matters; a clean PDF or Word document scans far better than an unusual format. Submitting a scanned image of text, rather than actual text, can produce a meaningless report. And a high similarity score is not automatically a problem. A correctly quoted and cited passage will still register as a match, because the words genuinely do appear elsewhere. The score flags overlap; a human decides what the overlap means.
How it works: the instructor's path
For instructors, the originality report is a lens, not a judge. After students submit, the instructor opens the assignment, sees the list of submissions, and finds a similarity indicator next to each one. Clicking it opens the full report.
A good report does three things. It gives an overall similarity percentage. It highlights the matched passages directly in the student's text. And it links each highlight to its source, whether that is a website, a journal article, or another student's paper. The reading happens in the highlights, not the number. A 40 percent score made entirely of a properly cited block quote and a standard methods section is fine. A 12 percent score that turns out to be one uncredited paragraph lifted word for word from a competitor's essay is the real problem. The percentage is where you start looking, never where you stop.
This is also where AI writing detection shows up, if the institution has it enabled. The instructor may see a second indicator estimating how much of the document reads as machine-generated. The same discipline applies, only more so. AI detection is probabilistic by nature, and false positives are a genuine risk, especially for multilingual students and for writing that happens to be plain and formulaic. We would never recommend acting on an AI flag alone. Treat it as a prompt to look closer, talk to the student, and compare the work against what you know of their voice from earlier in the term.
The strongest instructors use these tools before the grade exists, not after. Sharing the originality report with a student and asking them to explain a flagged passage turns a potential discipline case into a conversation about academic integrity. That conversation teaches far more than a points deduction ever will.
How it works: the administrator's path
Administrators rarely look at individual reports. Their job is to make sure the checks exist, run reliably, and respect privacy. In Brightspace, that work happens through an integration.
Most plagiarism and AI-detection tools connect to Brightspace through LTI, which stands for Learning Tools Interoperability. LTI is the standard handshake that lets an outside service appear inside the LMS as if it were native. An administrator registers the tool, supplies the credentials the vendor provides, and sets the scope of access. Newer integrations use LTI 1.3 and LTI Advantage, which tightened the security and data-sharing model compared with older versions. If you are setting this up today, you want the 1.3 path.
Once the tool is registered, the admin decides where it is available, which courses or departments, and what the defaults are. Should originality checking be on by default for every new assignment, or opt-in per instructor? Can students see their own scores? Are submitted papers stored in a shared repository for future comparison, or excluded? That repository question is more consequential than it looks. Storing student work strengthens future checks but raises real questions about consent and data ownership, and the right answer depends on your institution's policies and local law.
The other half of the admin role is communication. A plagiarism tool that instructors do not understand gets misused, and misuse erodes trust faster than almost anything else in academic integrity work. The most successful rollouts pair the technical setup with plain guidance: what the score means, what it does not mean, and how to talk to a student about a flag.
Common misconceptions
A few myths come up again and again, and clearing them up makes everyone's life easier.
"A high similarity score means cheating." No. It means overlap. Quotes, citations, bibliographies, and common phrases all register. The number is a starting point for human judgment.
"A zero score means the work is clean." Also no. A similarity tool only catches what it can compare against. Paraphrased ideas without attribution, work bought from a writer, or text the database has never seen can slip through. Originality checking lowers the odds of missing something; it does not guarantee it.
"The plagiarism check and the AI check are the same thing." They are not, and conflating them produces bad decisions. One measures overlap with existing sources. The other estimates the probability of machine generation. They answer different questions and fail in different ways.
"Brightspace does the detection itself." Brightspace provides the workflow. The analysis comes from an integrated service connected through LTI. Knowing this matters because it tells you who to call when something looks wrong.
The short version
Brightspace is the building, not the inspector. Students submit through the Assignments tool, instructors read the highlighted report rather than the headline number, and administrators wire up the integration through LTI and set the policies that govern it. The technology is genuinely useful, but only in the hands of people who treat a similarity score as the beginning of a question, never the end of one. Used that way, an originality check stops being a gotcha and becomes one more way to teach students how to write honestly.